Test method and test equipment for cable magnetic flux leakage test

Abstract

In a test method for performing a cable magnetic flux leakage test for detecting defects, a test device (100) and a cable (200) are moved relative to each other in a longitudinal direction of the cable. In this case, by means of a magnetizing device (120) of the testing device, the part of the cable is magnetized segment by segment in such a way that the magnetizing field line in the cable is oriented substantially in the longitudinal direction of the cable. The circumference of the magnetized portion of the cable is scanned by means of a magnetic field sensitive probe (140) of a probe device (130) for detecting a leakage magnetic field due to a defect. The probe device has a plurality of magnetic field-sensitive probes (140) which are arranged offset from one another in the circumferential direction and are arranged at a test distance (142) from the surface of the material under test during the test. The electrical probe signals of these probes are evaluated for qualification of defects. Each probe of the probe arrangement is arranged in a movably arranged test shoe (150) having a sliding surface (146) for sliding on the circumferential surface of the cable, the probe (140) being arranged at a probe distance (145) relative to the sliding surface. In a first test configuration, the test shoe (150) is pressed onto the circumferential surface such that the sliding surface (146) is in contact with the circumferential surface and the probe is held at a limited first test distance from the surface of the material under test substantially corresponding to the probe distance.

Description

Technical Field

[0001] The present invention relates to a testing method for performing a cable magnetic flux leakage test to detect defects and a testing device suitable for performing the testing method. Background Art

[0002] A typical cable, such as that used in cableways, elevators, or bridge construction, consists of several individual wires twisted together. The cable can be made directly from several individual wires or from several strands, each of which is made up of several individual wires. The raw material for the cable is usually ferromagnetic steel wire.

[0003] The flexibility of a cable stems from the twisting of individual conductors or strands of wire around each other, a process called lay. The lay directions of the strands and individual conductors can be the same or opposite. The distance a conductor or strand of wire travels at the same circumferential position is called the lay length.

[0004] Cables are often used in locations where the load-bearing capacity of ropes (e.g., plastic ropes) is insufficient, and where the flexibility of rods or tubes is insufficient. These locations are often outdoors, where the cables may be subject to severe weather conditions such as high and low temperatures, rain, snow, ice, wind, rain, and thunderstorms. In cableways in particular, the cables are subject to additional loads due to normal use. These loads include, for example, the constant or frequently repeated bending or clamping of the gondola.

[0005] In some cases, minor surface damage to cables, or even the breakage of individual conductors within the cable, can be tolerated or even repaired. However, cables must never be torn under any circumstances, as they are often a safety-relevant part of the system. Typical damage that can occur in cables includes: conductor breakage, crushing or deformation due to clamping, conductor dents, damage caused by the cable detaching from the conduit, surface damage caused by lightning strikes, and corrosion.

[0006] In order to ensure the operational reliability of cables, for example on cableways, bridges, elevators etc., regular inspections of the cables are required.

[0007] A proven method for inspecting cables for defects is the magnetic flux leakage test. To do this, sections of the cable are magnetized using a magnetic field with field lines oriented substantially in the longitudinal direction of the cable, using a magnetizing device in a testing machine. Simultaneously, a magnetic field-sensitive probe of a probe system scans the circumference of the magnetized section to detect any magnetic field leakage caused by defects.

[0008] Permanent magnets are typically used as magnets. These can be wrapped around the cable using a folding mechanism and magnetize the portion of the cable located between the poles of the magnetizing device as evenly as possible. Document DE 3821070 A1 shows an example of such a test device. This device includes a sensor arrangement with an inductive probe. The sensor arrangement consists of two coils (so-called half-shells) that wrap around the cable at 180° angles. The signals from these coils can be superimposed to form a fault signal. This probe arrangement is cost-effective and effective, particularly suitable for locating wire breaks. This is because the distance between the probe and the cable and the integration of the variation in magnetic flux over the entire circumference allow for good detection of internal cable damage. A disadvantage of this arrangement is that surface faults are poorly detected or even not detected at all. These defects must then be detected through regular visual inspections of the cable, which can often take several days for cables several kilometers long and is therefore costly.

[0009] Visual inspection requires trained personnel. Patent document EP 2383566 B1 discloses a test method for computer-assisted optical testing of a cable having multiple conductors or optical fibers using image processing based on an image dataset of at least a portion of the cable.

[0010] In some testing systems, the probe assembly includes more than two magnetic field-sensitive probes, which are staggered relative to each other in the circumferential direction. During testing, these magnetic field-sensitive probes are positioned at a limited test distance from the surface of the material being tested. The electrical probe signals from these probes are evaluated to characterize the defect.

[0011] Patent document EP 1995589 B1 discloses a relatively complex magnetic flux leakage testing device capable of detecting defects in the conductors of a cable even when these defects are located within the cable. The testing device comprises a magnetizing device and a plurality of magnetic detection devices. These magnetic detection devices include: a first magnetic detection device for detecting magnetic flux surrounding the entire circumference of the cable; a plurality of second magnetic detection devices; and a plurality of third magnetic detection devices. These second and third detection devices are arranged with gaps between them in the circumferential direction. There is also a total measuring device, to which the output of the first magnetic detection device is supplied and in which the total value of the leakage magnetic flux is measured; and a differential measuring device, to which the outputs of the second and third magnetic detection devices are supplied. There is also a memory; a first CPU, which is designed to receive signals from the differential measuring device and the total measuring device, and is designed to calculate the ratio between the signals from the differential measuring device and the total measuring device, and is designed to output a signal indicating the damage depth based on the relationship between the damage depth and the signal ratio previously stored in the memory; a second CPU, which is designed to calculate the damage extent based on the signal from the differential measuring device and based on the threshold data stored in the memory.

[0012] Here, the probe is spaced far enough from the cable that an eccentric position of the cable in the test apparatus and / or gross faults on the cable surface (such as protruding wires) do not damage the sensor (see, for example, US 5198765 A). Summary of the Invention

[0013] The object of the present invention is to specify a test method and a test device for magnetic flux leakage testing of cables, which test method and the test device are capable of reliably and highly selectively detecting faults of different fault types.

[0014] To achieve this object, the present invention provides a testing method having the features of claim 1 and a testing device having the features of claim 13. Advantageous developments are described in the dependent claims. The wording of all claims is incorporated into the description by reference.

[0015] This test method and this test device are used to detect defects in cables using a magnetic flux leakage test. The test device and the cable are moved relative to each other in the longitudinal direction of the cable. This can be achieved by keeping the cable stationary, i.e., not moving, while the test device is moved longitudinally along the cable. It is also possible to keep the test device stationary and move the cable longitudinally past the test device. In principle, it is also possible to move both the test device and the cable, but at different speeds and / or directions, so that the cable can be scanned over its entire length.

[0016] In this test method, sections of the cable are magnetized segment by segment using a magnetizing device of the test apparatus, so that the field lines of the magnetizing field, i.e., the magnetizing field lines, are oriented within the cable essentially in the longitudinal direction of the cable. The magnetizing device is preferably designed such that it surrounds the entire circumference of the cable, so that magnetic flux is introduced over the entire circumference and a uniform high magnetizing field is generated in the region of the test volume.

[0017] Using magnetic field-sensitive probes from a probe arrangement, the circumference of the magnetized portion of the cable is scanned to detect stray magnetic fields caused by defects in the cable. The probe arrangement comprises a plurality of magnetic field-sensitive probes, which are arranged staggered relative to one another along the circumference and, during testing, are positioned at a predetermined test distance from the surface of the material being tested. These probes generate electrical probe signals, which are evaluated to characterize the defect.

[0018] In the test method and the test apparatus, each probe of the probe device is arranged in a movably mounted test shoe having a sliding surface for sliding on the circumferential surface of the cable. The probe is arranged within the test shoe at a probe distance set back from the sliding surface.

[0019] In the first test configuration, during the relative movement between the test device and the cable, the test shoe is pressed onto the circumferential surface so that the sliding surface comes into contact with the circumferential surface and the probe is correspondingly maintained at a limited first test distance from the surface of the material to be tested, which first test distance substantially corresponds to the probe distance.

[0020] Here, the term "circumferential surface of the cable" refers to the cable's envelope, i.e., the smallest surface in contact with the outermost portion of the cable and bridging the depressions or valleys between the individual conductors or strands. This envelope can be, for example, a cylindrical surface or, if necessary, an elliptical cylindrical surface. The sliding surface of the test shoe is large enough to allow it to move along this envelope without significantly sinking into the valleys between the conductors or strands. This ensures relatively smooth operation or gentle sliding on cables without significant damage (e.g., the protruding ends of torn individual conductors).

[0021] This method differs significantly from previously known cable testing procedures. To the best of the inventors' knowledge, in conventional systems, the probe is spaced far enough from the cable that an off-center position of the cable within the testing apparatus and / or severe faults on the cable surface (such as protruding wires) do not damage the probe. This necessarily results in a relatively large testing distance, which can also vary over time if the cable's position within the probe assembly changes during travel.

[0022] In contrast, the procedure proposed herein provides a way to perform tests using a substantially constant test distance from the cable surface. Distance-dependent fluctuations in the signal amplitude are thereby substantially eliminated, making it possible to reliably attribute changes in the signal amplitude to possible defects. A constant test distance is crucial for high reproducibility and, therefore, high reliability of the test results, making them comparable with one another. Since the probe is set back relative to the sliding surface within the test shoe, the probe of the test shoe remains protected from mechanical contact with the surface. If the surface of the cable becomes damaged, for example due to a torn conductor protruding outwards, the test shoe's movable mounting allows it to deflect at the damaged point and thus pass through it without causing damage. A rise in the test shoe, for example due to a protruding conductor, can lead to a reduction in test sensitivity, particularly for faults close to the surface. However, this rise can be identified in various ways, allowing these points of the cable to be assessed as faulty.

[0023] A further advantage offered by the arrangement of the probe within the test shoe is that the probe can, if necessary, be arranged at a very small test distance from the circumferential surface of the cable. In a preferred embodiment, this test distance is in the range of 0.5 mm to 5 mm, in particular in the range of 1 mm to 3 mm, in the first test configuration. In other words, the distance of the probe to the sliding surface is preferably within these ranges. As a result, a high sensitivity is achieved for defects close to the surface, which would not be detectable in the probe signal when the magnetic field-sensitive probe is arranged at a greater distance from the surface. The small distance therefore results in a very favorable signal-to-noise ratio for defects close to the surface. This makes it possible, on the one hand, to identify defects that would be noticeable even during a visual surface inspection. Furthermore, with the magnetic flux leakage test, it is also possible to detect defects close to the surface that would escape optical inspection because they do not change the appearance of the surface.

[0024] The probes of the probe arrangement are preferably designed and arranged so that the cable can be continuously tested in the circumferential direction in a single pass. A single probe scans a test track extending substantially in the longitudinal direction of the cable, the width of which (test width) is determined by the effective width of the probe transverse to the scanning direction. The test tracks of all probes partially or completely overlap with the test tracks of the other probes, so that no sensitivity gaps form between the test tracks.

[0025] In order to achieve continuous testing in the circumferential direction on the one hand and fine position resolution in the circumferential direction for detecting defects close to the surface on the other hand, a preferred embodiment, such as one with a conventional cable diameter of 30 mm to 70 mm, provides that the probe arrangement has more than 20 probes distributed in the circumferential direction, preferably 30 or more probes, in particular 60 or more probes. The number of probes in the probe arrangement can even be in the order of 100, for example between 80 and 120. In addition to the achievable position resolution in the circumferential direction, this fine division in the circumferential direction has other advantages, which will be explained in conjunction with the other aspects.

[0026] To achieve high position resolution in the circumferential direction, it is also preferably provided that each of the probes has a test width, measured in the circumferential direction, corresponding to an arc length in the range of 1 mm to 5 mm, in particular 2 mm to 3 mm. The term "test width" here refers to the effective width of the probe's test track, but not the width of the probe itself. This width may be significantly smaller than the width of the test track. Preferably, the test width may be approximately 0.5 to 2 times the nominal sensor distance.

[0027] Preferably, the test device has three, four, or more test shoes distributed around the circumference and movable relative to one another. This allows the test device to adapt well to different cable cross-sectional dimensions, and the same test device can be used to test cables from a specific diameter range. According to the inventors' experience, three or four individually movable test shoes are sufficient for this purpose. While multiple test shoes could be provided, this would, in some cases, make the design of the test device unnecessarily complex.

[0028] Preferably, each test shoe is movably mounted on the body of the test device by means of a single-part or multi-part articulation. The articulation is a connection structure that is itself movable. The articulation should preferably be constructed so that: there is freedom of movement relative to the stroke axis in a radial plane containing the test shoe, while the freedom of movement transversely to the stroke axis (for example, tilting or large lateral deflections) does not exist or exists only slightly and may be subject to higher resistance. In this way, collisions between adjacent test shoes can be avoided in particular. The articulation can, for example, include a parallelogram guide and / or one or more solid hinges.

[0029] Each test shoe preferably has a plurality of probes arranged offset relative to one another in the circumferential direction. These probes preferably collectively cover a circumferential angular range greater than 360° divided by the number of test shoes, at least in a first test configuration. This ensures that the edge regions of circumferentially adjacent test shoes partially overlap, enabling continuous testing with varying diameters or ovality even in transition regions between adjacent test shoes. The test shoes can be constructed essentially identically, thereby keeping the costs for providing the test device moderate.

[0030] According to one embodiment, the test device has a controllable switching device for switching the test device between a first test configuration and a raised configuration, in which the test shoe is held in a raised position so that the sliding surface is arranged at a predeterminable distance from the surface or circumferential surface of the cable. The switching device can be actuated, for example, by remote control.

[0031] If necessary, a protective device can also be provided for detecting or identifying structural defects in the cable. This protective device can generate a control signal to then actuate the switching mechanism and cause a switch from the first test configuration to the raised configuration. For example, the evaluation can be performed such that if an excessively high signal is detected, the test shoe is raised by actuating the switching mechanism; or the test shoe or magnetization device can include at least one additional sensor that measures a force parallel to the direction of transport and triggers the raising mechanism if a limit value is exceeded. In some cases, a mechanical sliding ramp on the test shoe can also be sufficient.

[0032] Another advantage offered by the controlled switching approach is that testing can be performed in the raised configuration, with the probe positioned at a second test distance from the surface, where the second test distance is greater than the test distance in the first test configuration. The difference between the first and second test distances can be a multiple of the first test distance, for example, within a range of at least 5 mm, such as between 10 mm and 25 mm. Therefore, when testing using the second test configuration, the probe is positioned at a greater distance from the cable surface. This reduces sensitivity to defects close to the surface, and the probe signal generally only provides information about deeper defects. Thus, the test depth can be varied by switching between the first and second test configurations. More specifically, this configuration is sensitive to defects close to the surface at short distances and can also detect deep faults. As the distance increases, sensitivity to faults close to the surface decreases, ultimately reaching the same sensitivity as for deep faults.

[0033] In some cable testing variations, at least one pass is performed using a first test configuration and at least one pass is performed using a second test configuration. Thus, using the same test setup, reliable information can be obtained about larger flaws at any depth, where the effectiveness of the probe signal regarding the flaw's depth decreases with distance. If desired, information about smaller flaws on the cable surface can also be determined.

[0034] In some variants, a contact monitoring system is provided that automatically detects whether the test head has lost contact with the cable surface, for example due to a protruding conductor. If the distance to the test head becomes too great due to the loss of contact, the test is no longer reliable. The cable section passed during the loss of contact phase can then be marked as untested or faulty. To this end, some embodiments provide for the signal frequencies of all sensors of the test head to be continuously determined. If at least one sensor does not detect the high-frequency signal that is consistently generated due to a gap in the conductors, this indicates that the test head has lost contact with the cable surface.

[0035] The goal of cable testing is to determine as precise information as possible about the presence of defects on and in the cable and, in doing so, determine the location of the defects. According to one embodiment, for this purpose, special steps are performed when evaluating the probe signals. The evaluation of the probe signals comprises a number of mutually coordinated operations that are specifically adapted to the characteristics of the probe arrangement, which has the potential for high position resolution in the circumferential direction (in a first test configuration), and to the requirements for defect detection at different depths. According to one embodiment, the evaluation of the probe signals comprises a mapping operation, in which, for each probe signal, signal information representing the probe signal is associated with position information representing the location of occurrence of the probe signal to form position-dependent signal data; a matrix formation operation, in which the position-dependent signal data or signal data derived therefrom are stored in correctly assigned fields of a base matrix; and at least one evaluation operation, in which the position-dependent signal data from at least two adjacent fields of the base matrix in the evaluation direction are associated with one another using at least one evaluation algorithm.

[0036] In the mapping operation, for each probe signal, signal information representing the probe signal is associated with position information representing the location of occurrence of the probe signal in the scanned surface area. Position-dependent signal data is thereby generated. This mapping operation provides a unique association between signal information and position information, which can then be used as the basis for generating a "map" of the corresponding scanned surface area.

[0037] Then, in a matrix formation operation, the position-dependent signal data (or signal data derived therefrom) are stored in correctly assigned fields or elements of the base matrix. Unlike the common definition of a matrix in mathematics (a two-dimensional matrix arrangement of elements along rows and columns), the term "matrix" in this case refers to an n-dimensional arrangement of elements, where n is greater than or equal to 2. Therefore, a matrix for the purposes of this application can have more than two dimensions, such as three or four dimensions. In a broad sense, the matrix provides an association between position information, signal information, and possibly other information about parameters that affect the test method and its results. In one embodiment, the association is performed as follows:

[0038] The first dimension of the basic matrix represents signal information, which contains information about the leakage flux measured at a specific position on the cable surface. The signal information can exist as a scalar parameter (for example, the signal amplitude or a selected component of the leakage flux) or as a vector parameter. The second dimension of the basic matrix represents the position along the longitudinal direction of the cable. The third dimension of the basic matrix can represent the position along the circumferential direction of the cable. If a separate channel is assigned to each probe of the probe device, the position of the third dimension can also be given by the channel number.

[0039] This method variant then includes at least one evaluation operation in which position-dependent signal data from at least two adjacent fields along the evaluation direction of the base matrix are correlated with one another using at least one evaluation algorithm. This aspect will be explained in more detail in conjunction with the exemplary embodiments.

[0040] Preferably, bipolar signal information, ie the "raw" signal information from the unrectified probe signal, is used when forming the base matrix. Thus, information which would be lost due to rectification can also be used for the evaluation.

[0041] There are different ways to determine the position of a defect in the longitudinal direction of a cable. A prerequisite for this is that the travel period is known. For example, a circulating or measuring wheel can be used, which is fixedly mounted relative to the test device and runs along the cable. Alternatively, a proximity sensor can be used, which is oriented so that it can detect the individual twists of the cable passing by. In one embodiment, the travel period is calculated from the probe signal itself. To this end, during the evaluation, a periodically varying signal component of the probe signal is determined, which is correlated with the lay length of the cable, and the associated period length of this signal component is used to determine the axial position coordinates of the location of occurrence of the probe signal. A separate measuring device for measuring the travel period can thus be omitted.

[0042] Defects in cables can occur at different depths for different reasons, i.e., they can appear on or near the surface, or inside the cable, for example, near the center. In a preferred embodiment, the depth or depths to be checked can be specifically defined. Thus, the test can include detecting defects at a single depth or at multiple depths within the cable.

[0043] To this end, according to one embodiment, pre-filtering of the probe signal is performed. This pre-filtering includes depth-specific bandpass filtering for each depth position, preferably with adjustable cutoff frequencies. The lower cutoff frequency should be set to the lowest frequency of the probe signal expected at that depth position, and the upper cutoff frequency should be set to the highest frequency of the probe signal expected at that depth position. This utilizes the fact that defects close to the surface, when passing through the probe device, generate a relatively short signal in terms of time, which contains relatively high-frequency components. In contrast, central defects, i.e., defects in the center of the cable, generate a signal that extends longer in terms of time and contains correspondingly lower frequencies. Depth levels radially between the cable center and the surface correspondingly generate fault signals with medium-frequency signal components. Thus, the depth-specific bandpass filtering allows, to a certain extent, the focus of the test to be precisely defined.

[0044] In a preferred embodiment, another approach for specifically limiting the test to one or more specific depth positions and determining depth-position-specific signal information is used. During the evaluation, the probe signals of a specifiable probe group comprising two or more probes arranged offset in the circumferential direction are evaluated together using directly adjacent or partially overlapping test tracks, so that the probe group forms an effective probe having an effective test width in the circumferential direction that can be specified by the number of probes in the probe group. In other words, the probe signals of multiple adjacent test tracks can be integrated, and the resulting signal is then associated with a virtual probe length derived from the effective length of the adjacent probe group.

[0045] For example, to detect near-surface defects, i.e., defects located on or near the surface, the probe signals of a probe set comprising up to 10% of all probes in the probe arrangement (in particular, only two, three, or four of these probes) can be evaluated. Such active probes have a relatively narrow effective test width. On the one hand, this results in a relatively precise determination of the circumferential position of the defect, since the effective test width covers only a small portion of the circumference. For depth resolution, it is important to utilize a probe set with a relatively short virtual probe length, in particular to detect the high-frequency, short signals typical of near-surface defects, where their signal amplitudes stand out sufficiently clearly from the background signal.

[0046] If, on the other hand, one is looking for defects located in or near the center of the cable, it is preferable to evaluate the probe signals of a probe group that includes 90% or more of all probes of the probe arrangement, possibly all of the probes, so that effective probes are provided that completely surround the circumference and, by integrating the signals of the individual probes, provide a total signal similar to that of a single coil surrounding the cable.

[0047] To detect defects at medium depths, ie defects at a distance from the conductor surface and from the cable center, the probe signals of a probe group comprising more than 10% and less than 90% of all probes of the probe arrangement are evaluated accordingly.

[0048] An additional evaluation can determine how many adjacent probes identified the defect. If only one or a few probes detect a high-frequency signal, the defect is close to the surface. If all probes detect an increase in low-frequency magnetic flux leakage at the same longitudinal position along the conductor, the defect is located at or near the center of the cable. This allows for the determination of depth position information. This depth position information can be determined by measuring how many directly adjacent probes simultaneously detected a magnetic flux leakage signal exceeding a signal threshold associated with a specific defect.

[0049] According to an extension, the effectiveness of the test results can be further improved by performing a differential generation operation in which the difference of position-dependent signal information data of two matrix fields along a differential generation direction and at a differential distance from each other is determined.

[0050] The direction of differential generation is preferably set so that it substantially corresponds to the lay angle of the cable's strands or individual conductors. This allows for the elimination of magnetic flux leakage caused by the cable structure by differential generation of the helical or coil directions using a suitable differential reference. This allows for even more reliable detection of the difference between the signal of the defect being searched for and the background signal. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Further advantages and aspects of the invention emerge from the claims and from the description of exemplary embodiments of the invention, which are explained below with reference to the drawings.

[0052] Figure 1 shows a schematic cross section of an embodiment of a test apparatus when performing a magnetic flux leakage test on a passing cable;

[0053] Figure 2 The components of the probe assembly of the test equipment are shown in oblique perspective;

[0054] Figure 3 The components of the probe assembly of the test equipment are shown in an axial view and illustrate their relationship when detecting defects at different depths;

[0055] Figure 4 The diagram illustrates the various signal forms and related spectra of the probe signal as well as the adapted bandpass filtering;

[0056] Figure 5 Clarified the relationship regarding differential generation operations. DETAILED DESCRIPTION

[0057] Various embodiments of a test method and apparatus for magnetic flux leakage testing of cables are described below, based on examples. Magnetic flux leakage testing is used to detect defects on and / or in cables. Defects can include, for example, wire breaks, wire tears, unusual shape changes, or even damage closer to the surface, which may be accompanied by microstructural changes. Defects can be near the surface or internal to the cable.

[0058] The test is performed in travel, ie the testing device 100 and the cable 200 are moved relative to each other in the longitudinal direction of the cable, so that the entire length to be tested of the cable can be tested section by section by means of the magnetic flux leakage test.

[0059] Schematic Figure 1A portion of a cable 200 is shown, which consists of a plurality of individual conductors twisted together. In the example, the cable consists of a number of strands 210, which in turn consist of a plurality of individual conductors 205, which in the example are made of ferromagnetic steel. The cable can be provided for use in cableways, elevators, or bridge construction, for example. The cable has flexibility due to its multi-strand twisted structure. In the portion shown, the cable center 202, which extends in the longitudinal direction of the cable, extends in a straight line because the cable is subjected to axial stress. Cables of this type can be hundreds of meters or even thousands of meters long. The lay lengths of the strands and the individual conductors can be as shown, or they can be different. Along the longitudinal direction of the cable, the lay length 214 of the strands corresponds to the distance at which the strands are again at the same circumferential position. Correspondingly, the individual conductors also have a (smaller) lay length. Accordingly, the angle between the center direction of the cable (direction of the cable center 202 ) and the course of the litz wires is referred to as the lay angle 212 of the litz wires, and the corresponding situation applies to the lay angles of the individual conductors of the litz wires.

[0060] For reasons of operational reliability, most cables used for cableways, bridges, elevators, etc. require regular inspections for possible safety-related defects. For this purpose, in the exemplary case, a test device 100 is provided, Figure 1 , only some components of the test device are schematically shown. In the exemplary embodiment, the test device 100 is fixedly installed at a cableway station, the cable does not need to be removed from the system, and is moved relative to the test device 100 along its longitudinal direction. Alternatively, it is also possible that the cable does not move, but the test device is driven to move along the cable.

[0061] The test device 100 has a body 110, which can also be referred to as a housing, and carries the functional components shown on its inner side. The body has an opening 105 for a cable and can be, for example, in the form of a sleeve that can be folded open and closed in order to allow the cable to be introduced laterally into the interior of the housing or into the opening or to remove the test device.

[0062] In an ideal test configuration, the cable center 202 or neutral axis of the cable extends more or less coaxially with the channel axis 115 of the test apparatus defined by the body, where the channel axis extends centrally in the axially continuous channel opening.

[0063] The main body carries components of a magnetizer 120 on its interior, which is used to magnetize sections of the cable as it passes through the passageway opening. In the exemplary embodiment, the magnetizer 120 is equipped with permanent magnets 122-1 and 122-2 having north (N) and south (S) poles, and arranged such that the field lines 125 of the magnetizing field, at least in the region intermediate between the front and rear ends of the test apparatus, are oriented substantially in the longitudinal direction of the cable or parallel to the passageway axis. The magnetizer is designed so that it encompasses the entire circumference of the cable and introduces magnetic field lines or flux from all radial directions. This creates a uniform, high magnetizing field across the test volume. The main body facilitates magnetic return. Thus, in the exemplary embodiment, a constant axial magnetic field is used to magnetize the passing cable. Electromagnets, which require a cable for power supply, are generally unsuitable for mobile magnetizers, but can be used in stationary systems.

[0064] In the middle area between the front end and the rear end of the test device, components of the probe device 130 are arranged. Figure 2 Schematically shown in oblique perspective in Figure 3 The probe arrangement is shown in an axial view. The probe arrangement has a plurality of magnetic field sensitive probes 140 for detecting leakage magnetic fields caused by defects. These probes can be, for example, Hall probes. These probes are arranged staggered around the channel opening along the circumferential direction UR of the probe arrangement and maintain a radial distance from the channel axis 115. Figure 3 A section through the probe arrangement perpendicular to the channel axis is shown. It can be seen that the probe arrangement surrounds the cable, i.e., in such a way that the test tracks scanned on the outer surface of the cable with the individual probes are sufficiently wide that they overlap in the circumferential direction and thus ensure a continuous test over the entire circumference.

[0065] Furthermore, the structure is designed so that even small defects on or near the surface can be reliably detected using this probe arrangement. To this end, the individual probes are relatively narrow in the circumferential direction (for example, covering only an arc length range of 1 mm to 5 mm), and a large number of individual probes, for example 30 or more (32 in the example), are arranged circumferentially, although double or triple that number, or even more than 100, may be provided if necessary. In addition to the achievable position resolution in the circumferential direction, this fine division of the circumferential direction offers other advantages, which will be explained in detail later.

[0066] In order to perform a magnetic flux leakage test, each probe should be located at a defined (radial) distance from the surface of the object to be tested (here, a cable). This radial distance between the surface of the material to be tested and the respective probe is also referred to as a test distance 142. Typical test distances are usually in the range of about 2 mm, for example in the range of 0.5 mm to 5 mm, in particular in the range of 1 mm to 3 mm. In order to be able to maintain these relatively small test distances without risking a collision between the cable and the probe, which could damage the cable and / or the probe, each of the probes of the probe arrangement is arranged or mounted in a movably mounted test shoe 150.

[0067] Figure 1 The test shoe 150 is shown schematically in longitudinal section. Diametrically opposed test shoes are indicated only by dashed lines. Figure 2 An oblique perspective view is shown, which shows four arc-shaped test shoes 150-1 to 150-4, each of which carries multiple probes of the probe device. Each of the four test shoes extends over a circumferential angle range of approximately 100° to 110°. The test shoes are arranged in pairs diametrically opposite each other, and these pairs are arranged axially offset from each other in two planes. When viewed in the circumferential direction UR, the coverage areas of the test shoes overlap in an overlap area 155. The size of this overlap area ensures that the test shoes overlap at the ends even when they are positioned radially outward, so that no test gaps appear in the overlap area between circumferentially adjacent test shoes.

[0068] Each test shoe has, on the side intended to face the cable, a more or less flat or slightly convex cylindrical curved sliding surface 146, which is designed to slide on the outer surface of the cable 200 passing by, which moves relative to the test device, during the test. To this end, the test shoe is made of a metal or ceramic material with high mechanical hardness, at least on the side intended to face the cable. On the test shoe, a so-called upslope is formed in the passing direction, before and after the middle area with the probe. Figure 1 In the test configuration shown in FIG, the axial ends of these up-sloping surfaces are significantly set back from the magnets 122 of the magnetizer 120 in the radial direction, so that the passing wire will hit a portion of the slope anyway, but will not hit the end face of the test shoe.

[0069] The probe 140 is set back in radial direction relative to the associated sliding surface 146 of its pole shoe by the so-called probe distance 145. When the sliding surface slides on the surface of the cable in the test configuration, the probe distance 145 precisely defines the test distance 142, which can be maintained with high precision relative to the outwardly projecting apex of the strands 210 during the test.

[0070] The sliding surface of the test shoe slides on the surface and the probe is accordingly arranged at the smallest possible distance from the cable surface. This test configuration is also referred to here as the first test configuration. Here, the test distance 142 substantially corresponds to the probe distance 145.

[0071] The test shoes 150 are not fixed or rigidly mounted on the body, but rather have limited mobility in the radial direction, which is achieved by each of the test shoes 150 being movably mounted relative to the body by means of a suspension or articulation device 160 which is itself movable, so that the test shoe can be positioned at different distances from the channel axis 115. In the exemplary embodiment, a schematically illustrated articulation device 160 is provided, as well as a spring device 170, by means of which the test shoe 150 can be reliably pressed against the outside of the cable for testing. The articulation device can, for example, comprise a parallelogram guide, i.e., a mechanism by which the test shoe can be fixed on the arms and, during radial movement, maintained at an initial angle in a plane passing through the channel axis 115 (the sliding surface being parallel to the sheath line of the cable).

[0072] The test device is equipped with a lifting mechanism (not shown in greater detail) that allows for signal-controlled switching between the first test configuration shown and a raised configuration, in which the test shoe is in a raised position in which the sliding surface 146 is positioned at a predeterminable distance from the cable surface. This results in a second test distance for the test that is greater than the first test distance, corresponding to the probe distance. This allows for testing using at least two different test distances. The advantages of this feature will be explained later.

[0073] In the second test configuration, the test distance may be at least twice as large as in the first test configuration, the test distance may be, for example, at least 10 mm, for example in the range of 10 mm to 25 mm.

[0074] It may be expedient to perform at least one run with the first test configuration and at least one run with the second test configuration during cable testing. This allows significantly more important information to be determined about the defect state of the cable, in particular about the depth distribution of the defect.

[0075] During normal operation, continuous testing can be performed simultaneously at the second location. Then, at regular intervals or after events such as thunderstorms where the probability of surface failure is high, testing can be performed at the first location, for example, as an adjunct or supplement to the optical test. This test can also be configured to replace a visual inspection by an operator.

[0076] By switching between the first test configuration and the second test configuration in a controllable manner via a signal, not only the effectiveness of the entire test can be improved. More precisely, the operational reliability is also improved. In principle, it can happen that a single conductor in the outer area of ​​the cable tears off and protrudes outward from the surface. A collision with such a protruding end can cause the test to malfunction and, if necessary, damage or destruction of the test shoe or the entire test device. In some embodiments, a protective device is provided that can detect such structural defects and generates a signal when such a defect is detected, and the controller is configured so that, upon receiving this signal, a switch from the first test configuration to the raised configuration is automatically initiated. This makes a significant contribution to improving operational reliability.

[0077] The test shoe can, for example, be placed onto the object to be measured (cable) using a remotely controllable mechanism and automatically rise in the event of a hazard (e.g. multiple wires breaking from the surface of the cable). The hazard can be identified, for example, optically or in some other way, and a corresponding raising signal can be triggered. There are various ways of generating such a raising signal. In one case, the control unit or the evaluation unit is configured so that the test system recognizes an excessively high signal and then sends a raising signal to the test shoe. It is also possible that the test shoe includes an additional sensor which measures the force parallel to the transmission direction and which triggers the raising mechanism when a limit value is exceeded. However, a mechanical sliding ramp on the test shoe may also be sufficient to protect the test shoe.

[0078] The probe 140 is connected to an evaluation device 180 in a signal-conducting manner so that the evaluation device can process the probe signals. In this exemplary embodiment, the evaluation of the probe signals by means of the evaluation unit is specifically adapted to the characteristics of the probe device in order to achieve a high position resolution in the circumferential direction and to provide the possibility of assigning a defect signal to a defect at a specific radial depth.

[0079] To this end, the evaluation device is configured to perform a mapping operation, wherein, for each probe signal, signal information representing the probe signal is associated with position information representing the location at which the probe signal occurred. Position-dependent signal data is thereby generated. The signal data is further processed in a matrix formation operation designed to store the position-dependent signal data, or signal data derived therefrom, in correctly positionally assigned fields of a base matrix. Subsequent evaluation steps can then utilize the contents of the base matrix and use them for various analyses. At least one evaluation operation is performed in which, using at least one evaluation algorithm, position-dependent signal data from at least two adjacent fields of the base matrix along an evaluation direction are associated with one another.

[0080] The basic principles of matrix-based signal evaluation are described in US Patent No. 10082485 B2. With appropriate modifications and adaptations, this technology can also be used for cable testing.

[0081] The base matrix can have three or more dimensions. According to a suitable definition, the first dimension of the base matrix represents the signal information, that is, the information about the leakage flux measured at a specific position. Here, it is preferred to use bipolar signal information, that is, the raw signal information from the unrectified probe signal, so as to provide the complete information content of the local measurement results. For example, the position along the longitudinal direction of the cable (x value) can be selected as the second dimension, and the position in the circumferential direction (y value) (of the corresponding signal) can be selected as the third dimension. Therefore, in a broad sense, the base matrix provides an association between signal information and position information, and the base matrix can also contain other information if necessary.

[0082] In order to write the signals of the individual probes into a positionally reliable base matrix (x values ​​correspond to the axial position of the measured object, y values ​​to the circumferential position), the travel period is determined from the magnetic flux leakage signal itself. This is possible if information such as twist knots can be clearly identified in the magnetic flux leakage signal. Separate devices such as a recirculating wheel can be omitted. For some cables, such as those without twisted wires, a periodic signal with sufficiently high amplitude may not be available. In this case, a recirculating wheel or other suitable device can be used.

[0083] Next, the basic matrix or the signal information contained therein is filtered using different bandpass filters. The number of different bandpass filters depends on how many different depth positions in the cable need to be evaluated. For each depth position, an evaluation trajectory is introduced. For example, if three evaluation trajectories are required, namely one for the outer third of the cable (including the surface), one for the middle third and one for the innermost third of the cable (including the cable center), the basic matrix is ​​filtered for the outer third using a bandpass filter that allows high frequencies to pass, and for the innermost third using a bandpass filter for low frequencies. Correspondingly, the middle third uses medium frequencies. The use of bandpass filters for filtering the signal information is based on the understanding that the leakage magnetic signal produces a wider signal (usually also with a smaller amplitude) in the overtaking direction as the distance of the detected defect from the probe increases, and therefore has a lower frequency at the same overtaking speed.

[0084] Figure 4 This effect was elucidated. Figure 4Three superimposed subfigures show the bipolar signal form when plotting the signal amplitude A over time t on the left and the corresponding representation in frequency space (amplitude versus frequency t) on the right. The associated bandpass filters are represented by trapezoids. The three evaluation traces yield three different bandpass filter matrices through bandpass filtering.

[0085] For the next step, the integration length can be set on the evaluation device for each evaluation track. This integration length indicates how many adjacent probes in the circumferential direction the signal is to be summed. In this integration, for the same x value (same axial position) in the matrix, a sliding sum or sliding mean is formed over a certain number of y values ​​(corresponding to the individual probes recorded by the integration). Figure 3 Figure 1 shows three different integration lengths IL1, IL2, and IL3. Integration length IL1 represents integration over the entire circumference, i.e., the signals from all probes are summed. Integration length IL3 represents the shortest integration length for this example (with only three adjacent probes), while integration length IL2 represents a relatively large integration length in between, covering less than half and more than a quarter of the entire circumference.

[0086] In the maximum case (IL1), to detect a defect in the center conductor (DR1) of the cable, the sum is performed over the entire circumference, i.e., over all y values ​​at the x position of the matrix. The advantage of integrating over the entire circumference is that, in principle, the periodic flux variations between the individual conductors cancel each other out, resulting in a relatively low-noise sum signal, also called a reference signal.

[0087] For defects located on or close to the outer surface (e.g. surface conductor DR3), the integration length should be chosen to be relatively short, e.g. with only three probes side by side (see IL3). By means of high-frequency filtering, it is achieved that a probe further away may not detect the signal from a surface fault that is not directly beneath it. In addition, the signal from deeper defects is suppressed in this evaluation channel. Furthermore, when using high-frequency bandpass filtering, it may be helpful to adapt the angle of the integration direction to the lay direction of the cable in order to better capture the narrow fault maximum. For internal faults, the maximum of the fault signal is sufficiently wide in the circumferential direction that optimizing the integration direction does not bring advantages. The medium integration length IL2 is used to find defects on the internal conductor DR2 located between the cable center and the circumferential surface of the cable.

[0088] Furthermore, in a preferred embodiment, a differential located upstream of the integration can be applied in the differential generation operation. In the differential processing, the difference between adjacent values ​​is formed at a defined distance (differential reference). Thus, only those signal changes that deviate in the differential direction or whose spatial extension is less than the differential reference are retained. For cables, there are twist directions of the strands and conductors, which also produce leakage magnetic flux. Therefore, the differential direction should be consistent with the twist angle of the strands and conductors in order to suppress the inclined stripe pattern that appears in the cable. For ease of explanation, Figure 5 A schematic top view of a cable with the stranded conductors in the differential direction DF1 and the individual conductors in the differential direction DF2 is shown. The symbol OD indicates a small surface defect. The thick black stroke ZD and its two adjacent thick white strokes indicate a central conductor break, detectable over the entire circumference.

[0089] Differentiation is particularly important when the integration length does not cover the entire circumference of the cable. This is because, due to the circumferential symmetry of the cable structure, any magnetic flux leakage caused by the cable structure cancels out when integrated over the entire circumference, at least if the cable structure is completely uniform and all sensors are at the same distance from the cable along the circumference. For shorter integration lengths, which are chosen to detect faults close to the surface with high sensitivity, the magnetic flux leakage caused by the cable structure can be pre-cancelled by performing a helically extended differencing with a suitable differential reference (-> the width of the magnetic flux leakage in the direction of travel depends on the depth position).

[0090] These three operations (depth-specific bandpass filtering, integration, and differentiation) or the bandpass filtering, length and direction integration, and differential reference and direction algorithms used for this purpose can improve the signal-to-noise ratio of defects depending on depth by suppressing the periodic signals generated by individual strands and conductors. Furthermore, the depth-dependent signal level of a conductor break can be compensated by identifying the depth position.

[0091] The following summarizes some of the special features of an embodiment of cable testing. These features can be used individually or in combination. The test device has a magnetizer that, in the ready state, continuously surrounds the cable in the circumferential direction to generate a uniform magnetic field oriented in the longitudinal direction in the test space. The test device has test shoes that slide over the test object (cable), enabling a significantly reduced and uniform probe distance compared to conventional cable testing. Multiple magnetic field-sensitive sensors are mounted in each of these test shoes to continuously scan the leakage magnetic flux over the entire circumference of the test object with high resolution and accuracy. Various bandpass filters can be used, depending on the depth of the fault. The integration length can be selected based on the depth of the fault. The integration and differential directions can be adapted to the cable's lay length. The differential reference can be adapted to the depth of the fault. Depth-dependent signal levels can be compensated for conductor breaks. The travel period can be calculated from the leakage magnetic flux signal, for example, using the lay length and / or a correlation function of the probes in the overlap region of the test head. An automatic lifting device can be provided.

Claims

1. A method for conducting a cable magnetic flux leakage test to detect defects, wherein: the test device and the cable are moved relative to each other in the longitudinal direction of the cable, magnetizing sections of the cable segment by segment by means of a magnetizing device of the testing device such that the magnetizing field lines in the cable are oriented substantially in the longitudinal direction of the cable, By means of a magnetic field sensitive probe of a probe device, the circumference of the magnetized portion of the cable is scanned to detect leakage magnetic fields caused by defects, wherein the probe device has a plurality of magnetic field sensitive probes, which are arranged staggered with each other in the circumferential direction and are arranged at a certain test distance from the surface of the material to be tested during the test, and evaluating an electrical probe signal of the probe for characterizing the defect, It is characterized by: Each probe of the probe device is arranged in a movably mounted test shoe, the test shoe having a sliding surface for sliding on the circumferential surface of the cable, wherein the probe is arranged set back a certain probe distance relative to the sliding surface, and In a first test configuration, the test shoe is pressed onto the circumferential surface so that the sliding surface is in contact with the circumferential surface and the probe is held at a limited first test distance from the surface of the material under test, substantially corresponding to the probe distance.

2. The testing method according to claim 1, wherein: The test distance in the first test configuration is in the range of 0.5 mm to 5 mm, in particular in the range of 1 mm to 3 mm.

3. The testing method according to claim 1 or 2, characterized in that: The test device has three or four or more test shoes distributed in the circumferential direction and movable relative to each other, wherein, preferably, each test shoe has a plurality of probes arranged staggered with each other in the circumferential direction, and the probes preferably jointly cover a circumferential angle range greater than 360° divided by the number of the test shoes at least in the first test configuration; and / or couple magnetic flux from all radial directions into the cable; and / or perform continuous testing in the circumferential direction; and / or the magnetization device is designed so that the magnetization device seamlessly surrounds the entire circumference of the cable in the operating state.

4. The test method according to any one of the preceding claims, characterized in that a controlled switch between the first test configuration and a raised configuration, in which the test shoe is held in a raised position so that the sliding surface is arranged at a prescribable distance from the circumferential surface of the cable, wherein, preferably, a protection device is present for detecting structural defects on the cable and for generating a control signal for automatically initiating the switch from the first test configuration to the raised configuration.

5. The test method according to any one of the preceding claims, characterized in that Testing is performed in a raised configuration, wherein the probe is arranged at a second test distance from the surface, wherein the second test distance is greater than the test distance in the first test configuration, preferably at least 5 mm greater, wherein, preferably, when testing the cable, at least one stroke is performed using the first test configuration and at least one stroke is performed in the second test configuration.

6. The test method according to any one of the preceding claims, characterized in that Automatic contact monitoring, which automatically detects whether the test head has lost contact with the cable surface, wherein, preferably, for the contact monitoring, the signal frequencies of all sensors of the test head are continuously determined and a contact loss signal is generated if at least one sensor does not detect a high-frequency signal which is always generated due to a wire gap on the cable surface, wherein, preferably, the cable section passed during the contact loss phase is marked as not tested or as defective.

7. The test method according to any one of the preceding claims, characterized in that The evaluation of the probe signal comprises the following steps: a mapping operation in which, for each probe signal, signal information representing the probe signal is associated with position information representing an occurrence position of the probe signal to form position-related signal data; a matrix formation operation in which the position-dependent signal data or signal data derived therefrom are stored in positionally correctly assigned fields of a base matrix; and at least one evaluation operation in which position-dependent signal data from at least two adjacent fields in an evaluation direction of the base matrix are correlated with one another using at least one evaluation algorithm, Preferably, bipolar signal information is used when forming the basic matrix.

8. The test method according to any one of the preceding claims, characterized in that During the evaluation, a periodically varying signal component of the probe signal is determined which is correlated with the lay length of the cable, and the axial position coordinates of the occurrence location of the probe signal are determined using the associated period length of the signal component.

9. The test method according to any one of the preceding claims, characterized in that The test comprises detecting defects at a single depth position or at a plurality of different depth positions of the cable; and performing pre-filtering of the probe signal, wherein the pre-filtering comprises for each depth position a depth-position-specific band-pass filtering, preferably with settable limit frequencies, wherein, in particular, the lower limit frequency is set to the expected lowest frequency of the probe signal for the depth position and the upper limit frequency is set to the expected highest frequency of the probe signal for the depth position.

10. The test method according to any one of the preceding claims, characterized in that During the evaluation, probe signals of a specifiable probe group having two or more probes arranged staggered in a circumferential direction are evaluated together using test tracks that are directly adjacent to or partially overlap each other, so that the probe group forms an effective probe having an effective test width in the circumferential direction that can be specified by the number of probes in the probe group.

11. The test method according to any one of the preceding claims, characterized in that For detecting defects located on or near the surface, the probe signals of a probe group comprising a maximum of 10% of all probes, in particular only two or three or four probes, are evaluated; and / or for detecting defects located in or near the center of the cable, the probe signals of a probe group comprising 90% or more, in particular all, of the probes of the probe arrangement are evaluated; and / or for detecting defects at medium depths, the probe signals of a probe group comprising more than 10% and less than 90% of all probes of the probe arrangement are evaluated; and / or it is determined how many directly adjacent probes simultaneously detect a leakage magnetic signal exceeding a signal threshold associated with a specific defect; and the depth position information of the defect is determined therefrom.

12. The test method according to any one of the preceding claims, characterized in that A differential generation operation, in which the difference of signal information data related to the positions of two fields along a differential generation direction and at a certain differential distance from each other is determined, wherein, preferably, the differential generation direction is set so that the differential generation direction basically corresponds to the winding angle of the twisted wire or the single wire.

13. A testing device (100) for performing a magnetic flux leakage test on a cable (200) to detect defects, wherein: The test device has a passage opening (105) for the cable, and the cable and the test device are movable relative to each other in the longitudinal direction of the cable, the test device comprising: a body (110) forming the passage opening (105) and defining a passage axis (115) located in the passage opening; magnetizing means (120) carried by the body for magnetizing a portion of the cable (200) as it passes through the passage opening (105) such that magnetizing field lines (125) in the cable are oriented substantially in the longitudinal direction of the cable; A probe device (130) having a plurality of magnetic field sensitive probes (140) for detecting leakage magnetic fields caused by defects, wherein the probes are arranged staggered with respect to one another in a circumferential direction around the passage opening (105); and an evaluation unit (180) for evaluating the electrical probe signal of the probe for qualitative determination of the defect, It is characterized by: Each probe of the probe device is arranged in a test shoe (150), and the test device has three or more test shoes, wherein each of the test shoes is arranged in a movable manner relative to the main body (110) so that the test shoe can be placed at different distances from the passage opening (115), A sliding surface (146) is provided for sliding on the circumferential surface of the cable (200), wherein each probe of the test shoe is arranged to be set back by a probe distance (145) relative to the sliding surface, and In a first test configuration, the probe (140) can be pressed onto a circumferential surface of a cable (200) passing through the passage opening so that the sliding surface (146) contacts the circumferential surface and the probe (140) is maintained at a limited first test distance (142) from the surface of the cable, substantially corresponding to the probe distance.

14. The testing device according to claim 13, characterized in that At least one, preferably multiple or all of the following conditions are met: The probe device (130) has more than 20 probes distributed in the circumferential direction, wherein, preferably, the probe device has 30 or more or 60 or more probes; Each of the probes (140) has a test width measured in a circumferential direction, the test width corresponding to a circumferential angle in the range of 5° to 10°; The magnetization device (120) is designed so that it surrounds the entire circumference of the cable in an operating state; The probe (140) of the probe device (130) is designed and arranged so as to enable continuous testing of the cable in the circumferential direction in a single stroke.

15. The test device according to claim 13 or 14, characterized in that At least one of the following characteristics: A test shoe (150) is movably mounted on the body (110) by means of an articulation device (160), wherein the articulation device preferably comprises a parallelogram guide and / or a solid hinge; A pressure device is provided, which preloads the test shoe in the direction of the channel axis, wherein the pressure device preferably comprises a spring device (170) with at least one spring.

16. Testing device according to any one of claims 13 to 155, characterized in that The evaluation device (180) is configured to perform an evaluation of the probe signal according to the features of the characterizing portion of any one of claims 6 to 12.

Images