Test method and test apparatus for stray flux testing

Abstract

The invention relates to a test method for stray flux testing of ferromagnetic test object (110), in particular ferromagnetic pipes, in order to detect defects, in which, in a continuous process, the test objects are conveyed in a direction of passage (102), first through a first subsystem (200) and then through a second subsystem (300) of a test apparatus (100) and are demagnetised before exiting the test apparatus. The first subsystem (200) is designed as a transverse defect testing system and has a first magnetising device (220) which generates a magnetisation in the test object, the field lines of which run substantially in the longitudinal direction of the test object. The second subsystem (300) is designed as a longitudinal defect testing system and has a second magnetisation device (320) which generates a magnetisation in the test object, the field lines of which run substantially in the circumferential direction of the test object. In each of the subsystems, a surface of the test object is scanned by means of at least one magnetic field-sensitive probe (250, 350) to detect stray magnetic fields caused by defects in the region of a test zone, and electrical probe signals from the probes are evaluated to qualify the defects. After passing through the test zone (230) of the first subsystem (200), the test object is largely or completely demagnetised on exiting the first subsystem before entering the second subsystem (300) by means of a DC magnetic field of a demagnetisation coil (400) assigned to the first subsystem, which field is directed opposite to the magnetic field of the first magnetisation device.

Description

[0001] Test method and test device for flux leakage testing

[0002] FIELD OF APPLICATION AND STATE OF THE ART

[0003] The invention relates to a test method for flux leakage testing of ferromagnetic test material for detecting defects according to the preamble of claim 1 and to a test device suitable for carrying out the method according to the preamble of claim 7.

[0004] Magnetic flux leakage methods are an important component in the non-destructive testing of semi-finished and finished parts for defects, both during the manufacturing process and during the cyclical inspection of finished parts. Magnetic flux leakage methods are less sensitive to some interfering material properties, such as surface roughness or scale in hot-rolled products, than, for example, eddy current or ultrasonic testing. This results in a better ratio between the useful signal and the interference signal (N / S ratio), enabling more reliable defect detection.

[0005] In a flux leakage test, a test volume of the test specimen is magnetized using a magnetization device and scanned using at least one magnetic-field-sensitive probe (flux leakage probe) to detect magnetic stray fields caused by the defects. This involves a relative movement between the probe and the surface of the test specimen.

[0006] The magnetic flux or magnetic field generated by a magnetizing device in the test material is spatially distributed essentially homogeneously in the defect-free material. In this case, no significant magnetic field gradients occur in the areas close to the surface. Cracks and other defects, such as blowholes, inclusions, or other discontinuities such as weld seams, etc., act as areas of increased magnetic resistance, so that field components near a defect are diverted around the defect and forced out of the metal into the area close to the surface. The forced-out field components are detected in the flux leakage method to detect defects. In a flux leakage measurement, a defect is detectable if the field components displaced from the test object reach the area of ​​the probe and cause a change in the field there sufficient for detection.The effect of expelling field components is particularly pronounced, and the test sensitivity correspondingly high, when the magnetization direction is oriented perpendicular to the main direction of extension of a defect. In other words, the magnitude of leakage flux signals from a defect decreases the more the defect deviates from the (ideal from the perspective of testing) orthogonal position to the main direction of magnetization. If the main direction of extension of a defect is aligned parallel to the longitudinal axis of the test material, these are referred to as longitudinal defects. To detect these easily, the magnetization direction should be oriented as perpendicular as possible to the main direction of extension, particularly in the circumferential direction of the test material.Transverse defects, i.e., defects whose main extension direction is aligned in the circumferential direction of the test material and thus perpendicular to the longitudinal axis of the test material, are particularly easy to detect when the magnetization direction is parallel to the longitudinal axis of the test material. Defects whose main extension direction is oriented obliquely to the longitudinal direction and obliquely to the circumferential direction (oblique defects) are more easily detectable the steeper the magnetic flux lines encounter the defect.

[0007] In order to detect longitudinal defects, transverse defects, and oblique defects, testing devices of the type considered in this application have two subsystems that are connected in series with respect to the direction of travel of the test material. In a test line, the test line first passes through a first subsystem, which is designed as a transverse defect testing system and accordingly causes magnetization of the test material in its longitudinal direction. This is followed by a second subsystem, which is designed as a longitudinal defect testing system and accordingly causes magnetization of the test material in its circumferential direction. Oblique defects can generate defect signals in both subsystems, the signal amplitude of which depends not only on the size of the defect but also on the angular position of the defect in relation to the longitudinal and circumferential directions.

[0008] A key quality criterion for flux leakage testing devices is reproducibility. This means that comparable defects must be detected with a comparable signal amplitude. Test devices often have to demonstrate, during customer acceptance, that they can detect comparable defects with a reproducibility of better than 3 dB. The inventors' experience shows that these strict criteria, especially for angular defects, cannot always be met, or only with considerable technological effort.

[0009] After passing through the subsystems, residual magnetization may remain in the test material, which can interfere with further processing steps or applications. For this reason, conventional testing devices or stray flux lines usually also include a demagnetization coil to demagnetize the tested test material after it has passed through the first and second subsystems. The demagnetization coil is usually a simple air coil, i.e. it consists of one or more copper windings without flux conductors or concentrators. The copper windings are operated with an alternating current, the frequency and amplitude of which are adjusted based on the diameter, wall thickness if applicable, and transport speed of the test material. Demagnetization using an alternating magnetic field is described, for example, in US Pat. No. 6,037,767.

[0010] TASK AND SOLUTION

[0011] Against this background, it is an object of the invention to provide a testing method and a testing device for flux leakage testing of ferromagnetic test material, which enable reliable testing for defects of different defect types, ie not only for longitudinal defects and transverse defects, but also for oblique defects, with good reproducibility.

[0012] To achieve this object, the invention provides a testing method having the features of claim 1 and a testing device having the features of claim 7. Advantageous further developments are specified in the dependent claims. The wording of all claims is incorporated into the description by reference.

[0013] A testing method according to the claimed invention is designed for flux leakage testing of ferromagnetic test material in a continuous process and serves to detect defects in the test material. The test material can be, in particular, a ferromagnetic tube, but possibly also a solid rod made of ferromagnetic material.

[0014] During the testing process, the test sample is conveyed along a passage direction, first through a first subsystem and then through a second subsystem of a test device. Before exiting the test device, it is demagnetized to a tolerable residual magnetization using a demagnetization device. The term "demagnetization" here means that the test sample either exhibits no residual magnetization at all after passing through the test device, or the residual magnetization is so low according to a specification that it does not interfere with further processing of the test sample.

[0015] The first subsystem is designed as a transverse defect testing system and has a first magnetization device that generates a magnetization in the test material whose field lines run essentially in the longitudinal direction of the test material. The second subsystem, downstream of the first subsystem, is designed as a longitudinal defect testing system and has a second magnetization device that generates a magnetization in the test material whose field lines run essentially in the circumferential direction of the test material. In each of the subsystems, a surface of the test material is scanned using at least one magnetic field-sensitive probe (flux leakage probe) to detect stray magnetic fields caused by defects in the area of ​​a test zone. The test zone lies between the inlet and outlet ends of the subsystem. Electrical probe signals from the probes are evaluated to qualify the defects.

[0016] A special feature of this test method compared to conventional solutions is that the test specimen, after passing through the test zone of the first subsystem, is largely or completely demagnetized upon exiting the first subsystem and temporally and spatially before entering the second subsystem by means of a DC magnetic field of a demagnetizing coil assigned to the first subsystem, which is directed counter to the magnetic field of the first magnetizing device. Preferably, a single demagnetizing coil is used. Multiple demagnetizing coils can be provided, particularly if other components are to be reduced in addition to the magnetizing component acting in the longitudinal direction.

[0017] For this purpose, the testing device is equipped with a demagnetization device which has a direct current source and a demagnetization coil connected thereto, which is arranged in a run-out region of the first subsystem and can be operated with direct current in such a way that the test material magnetized by the first magnetization device can be demagnetized completely or down to a tolerable residual magnetization by reducing the magnetization component running in the longitudinal direction of the test material.

[0018] The demagnetization coil arranged in the outlet region of the first subsystem thus generates a DC magnetic field whose polarity is opposite to the DC magnetic field of the first magnetization device. Therefore, this DC field demagnetization coil is also referred to in this application as a counter-coil. It can also be referred to as a compensation coil, since the magnetization generated in the test specimen by the first magnetization device can be partially or completely compensated with the aid of the DC counter-field, so that the test specimen exhibits no residual magnetization or only a tolerable residual magnetization upon entering the second subsystem. The claimed invention is based on several findings of the inventors. A first aspect concerns the sequence of the two subsystems.Until now, it was assumed that it was fundamentally irrelevant whether, when connecting the two subsystems in series (transverse defect testing and longitudinal defect testing), the transverse defect testing was carried out first, followed by the longitudinal defect testing, or whether the order was reversed. According to the claimed invention, however, a selection is made in this regard such that the transverse defect testing is necessarily carried out first, in which magnetization directed in the longitudinal direction of the test material is generated in the test material using the first magnetization device. This magnetization is then specifically eliminated using a DC magnetic field immediately upon the test material exiting the first subsystem, or is reduced to such an extent that the test material is completely demagnetized or enters the second subsystem (longitudinal defect testing) with only a tolerable residual magnetization.

[0019] As a result, the residual magnetization in the longitudinal direction, which is particularly critical with regard to residual magnetization, is reduced to a non-critical level immediately upon exiting the first subsystem, so that the downstream longitudinal defect test essentially proceeds in the same way as in a fresh, completely unmagnetized test piece. Since the longitudinal defect test requires magnetization generated by the second magnetization device with magnetic field lines in the circumferential direction of the test piece, the level of residual magnetization in the longitudinal direction remains low upon passing through the second subsystem and can even be further reduced, so that it is generally unnecessary to connect a demagnetization coil downstream of the second subsystem to eliminate the magnetization component directed in the longitudinal direction of the test piece.Since no space needs to be taken into account for a demagnetization coil either in front of or behind the second subsystem, the length of the second subsystem, which is critical for material guidance, can be kept short.

[0020] Further advantages arise from the fact that the DC-operated demagnetization coil is arranged in a run-out area of ​​the first subsystem, i.e., in the immediate vicinity of the first subsystem. This spatial proximity ensures that no, or at most very narrow, air gaps remain between the demagnetization coil and the magnetically conductive components of the first subsystem, so that only a small magnetic resistance needs to be overcome in this area. This means that the flux conductors present in the first subsystem can also be used to conduct the DC magnetic field generated by the demagnetization coil. Thus, the demagnetization coil does not require its own flux conductors, allowing the use of a demagnetization coil with minimal winding space.In addition, the demagnetization coil can have a high efficiency due to its spatial proximity to the first subsystem, since it can operate with a lower electrical power to still generate a counter field of sufficient strength.

[0021] According to a further development, the demagnetization coil of the demagnetization device is attached to a housing of the first subsystem. This allows for the greatest possible proximity to the first subsystem, and magnetically conductive components of the first subsystem can also be used to conduct the demagnetization field. Furthermore, the demagnetization coil can be installed and adjusted together with the first subsystem.

[0022] In some embodiments, the first magnetization device (magnetization device of the first subsystem) has a magnetization coil on each of the inlet and outlet sides, as well as test material passage nozzles made of ferromagnetic material, which are coupled via magnetic return conductors. The demagnetization coil is arranged on the first subsystem in such a way that field lines of a demagnetization field generated by the demagnetization coil run through the flux conductors on the outlet side of the first subsystem.

[0023] According to a further development, the direct current source is designed as an adjustable direct current source so that the strength of the demagnetizing field can be adapted to the respective requirements.

[0024] In preferred embodiments, a control loop is provided for automatically controlling the strength of the DC magnetic field generated by the demagnetizing coil as a function of a residual magnetization value, which can be determined in a residual magnetization detection operation. Thus, the strength of the compensating opposing field can be optimally adjusted to the requirements without operator intervention.

[0025] According to another formulation, a residual magnetization determination operation is provided, wherein a residual magnetization value is determined on the test piece after it has passed through the first subsystem and the demagnetization coil, said residual magnetization value representing the strength of a magnetization component directed in the longitudinal direction of the test piece. The strength of the DC magnetic field generated by the demagnetization coil is then automatically controlled as a function of the determined residual magnetization value. A separate measuring device could be provided to determine the residual magnetization value. In preferred embodiments, however, the already existing second subsystem is used as a measuring system for determining the residual magnetization value during the residual magnetization determination operation. The second subsystem can thus be used as a functional part of a control loop for setting the correct demagnetization field strength.

[0026] According to a further development, the second subsystem comprises a rotating head having pole pieces of the second magnetization device aligned radially to the test piece surface at diametrically opposite locations such that a constant magnetic field (rotating around its longitudinal axis) can be generated inside the test piece, the field lines of which run in the circumferential direction of the test piece, with at least one probe (leakage flux probe) arranged in the circumferential direction between each pole piece. Using such a second subsystem, a skew error calibration procedure can be performed, in which the strength of the constant magnetic field generated by the demagnetization coil is adjusted such that a skew error, regardless of its inclination (angle and inclination direction), generates essentially the same error signal amplitude in each probe scanning the skew error.

[0027] This unusual approach is based on the finding that residual magnetization present in the test material with a magnetization component in the longitudinal direction of the test material causes an oblique defect of a certain size to generate error signal amplitudes of varying strengths depending on its orientation. This orientation dependence exists as long as a significant magnetization component is present in the longitudinal direction of the test material. If, however, the magnetization component acting in the longitudinal direction is eliminated by appropriate demagnetization, an oblique defect of a certain size exhibits the same error signal amplitude regardless of whether it is a left-handed or a right-handed oblique defect.

[0028] This effect can be used to regulate or adjust the strength of the demagnetizing DC field generated by the demagnetizing coil. If the demagnetizing field strength is correctly adjusted so that the longitudinal magnetization component practically disappears, optimal reproducibility of skew errors is also achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Further advantages and aspects of the invention emerge from the claims and from the description of embodiments of the invention, which are explained below with reference to the figures.

[0030] Fig. 1 shows a schematic sectional view with components of an embodiment of the testing device;

[0031] Fig. 2 shows schematically a test specimen with a cylindrical coordinate system;

[0032] Fig. 3 shows schematically different types of errors;

[0033] Figs. 4A and 4B show in 4A different superpositions of magnetization field components and in 4B the effects on the error signal amplitudes of different error types in a test specimen without residual magnetization in the longitudinal direction;

[0034] Figs. 5A and 5B show in 5A different superpositions of magnetization field components and in 5B the effects on the error signal amplitudes of different error types in a test specimen with residual magnetization in the longitudinal direction.

[0035] DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] Embodiments of the claimed invention are explained below using a testing device 100 configured for a flux leakage test of ferromagnetic test material in the form of ferromagnetic tubes in a continuous process. The tubes can be, in particular, hot-rolled tubes, but possibly also welded or cold-drawn tubes. Fig. 1 shows a schematic sectional view with important components of the embodiment.

[0037] To describe pipes and their defects, a cylindrical coordinate system with the three coordinates r, (p) and z is used instead of a Cartesian coordinate system (see Fig. 2 and Fig. 3). Here, r is the radial coordinate, (p) the angular coordinate and z the height above the plane spanned by r and (p). The length of the pipe in the axial direction is described with the z-coordinate, the diameter and the wall thickness with the r-coordinate and the circumference of the pipe can be described with the cp-coordinate.

[0038] The inspection device 100 is designed for the detection of defects, imperfections, or discontinuities of various types and can, for example, reliably detect rolling defects on both the inside of the pipe (internal defects) and the outside of the pipe (external defects). Both longitudinal defects (LF) (defects with a main extension direction parallel to the pipe's longitudinal axis) and transverse defects (QF) (defects with a main extension direction in the circumferential direction or perpendicular to the pipe's longitudinal axis) and oblique defects (SF) (transverse to the longitudinal direction and the circumferential direction) can be reliably detected and characterized.

[0039] Longitudinal faults (LF) are oriented in the z-direction in cylindrical coordinates, while transverse faults (QF) are oriented in the cp-direction. Faults that are not oriented along the z-direction but also in the positive or negative cp-direction are referred to as skew faults. Right-handed skew faults (RH) have a positive cp component according to the definition used here, and left-handed skew faults (LH) have a negative cp component with an increasing z-component.

[0040] The testing device 100 operates in a continuous process. It comprises devices for conveying the test material 110 along a passage direction 102 through a first subsystem 200 and a second subsystem 300 of the testing device 100, which is arranged downstream of the first subsystem in the passage direction. The first subsystem 200 is designed as a transverse flaw testing system and has a first magnetization device 220, which is configured to generate a magnetization in the test material whose field lines run essentially in the longitudinal direction of the test material (z-magnetization, longitudinal magnetization). The second subsystem 300 is designed as a longitudinal flaw testing system and has a second magnetization device 320, which is configured to generate a magnetization in the test material whose field lines run essentially in the circumferential direction of the test material (cp-magnetization).

[0041] Each of the subsystems has a test zone 230, 330 with at least one magnetic-field-sensitive probe 250, 350, with which a surface of the test material can be scanned to detect magnetic stray fields caused by defects. The electrical probe signals from the probes are evaluated using an evaluation device 500 to qualify the defects.

[0042] The magnetic flux leakage test for transverse defects requires a magnetic flux that runs as parallel as possible to the pipe axis (z-direction), i.e., perpendicular to the defect direction. This magnetic flux, running longitudinally to the pipe axis, is generated in the example using two coils 222-1, 222-2 surrounding the pipe, which are spaced apart from each other at a distance approximately equal to their diameters. The coils of the first magnetization device 220 are connected to a direct current source (DC power supply) 224 and, during operation, are traversed in the same direction by a current of several amperes, which is sufficient to magnetize a ferromagnetic pipe so that an operating point suitable for the flux leakage test can be reached. Magnetic flux conductors, e.g.Nozzles made of ferromagnetic material adapted to the pipe diameter to be tested (inlet nozzle 225-1, outlet nozzle 225-2), as well as return conductors 227, which are located outside the coils, can minimize the resistance for the magnetic flux, which leads to higher flux densities in the material at the same current strength.

[0043] As a pipe section enters the first subsystem 220, the magnetic flux is directed into the pipe through the inlet nozzle 225-1. Between the nozzles, the ferromagnetic pipe (test specimen 110) conducts the magnetic flux to the outlet nozzle 225-2. There, the flux exits the material and is directed back through the outlet nozzle and the return conductor 227, closing the magnetic field lines.

[0044] In the center between the nozzles lies the test zone 230 of the first subsystem. There, test shoes are located, each containing a single or multiple (e.g., up to several dozen) sensors or probes 250. The test shoes and probes are arranged so that the pipe can be scanned over its entire circumference. The test shoes glide over the pipe, which is transported through the subsystem by driven rollers. At the location of the test shoes, the pipe has a magnetization as uniform as possible in the z-direction.

[0045] In the presence of a transverse fault, the pipe's cross-section is locally reduced. This results in a redistribution of the magnetic flux within the pipe, which also forces a small portion of the flux out of the pipe. This flux change at the pipe surface is detected by the sensors (probes) 250 and converted into a high-resolution leakage flux fault image using suitable algorithms.

[0046] After a pipe section has been tested and the magnetic flux has jumped from the pipe into the outlet nozzle 225-2, the pipe remains magnetized. The level of residual magnetization or remanence depends on the material and the level of magnetization. The direction of the residual magnetization corresponds to the magnetization direction of the first subsystem, i.e., the longitudinal direction of the test material or z-direction. The test for longitudinal defects is carried out after the transverse defect test using the second subsystem 300, which is located downstream of the first subsystem 200 at a certain distance in the direction of travel. The second subsystem 300 has a rotating head rotatable about the travel axis. The second magnetization device comprises two coils 322-1, 322-2, each wound on a core and arranged around the pipe in opposite radial directions (r-direction).On the side facing the tube, each core has a pole piece 325-1, 325-2 adapted to the tube diameter (see, for example, EP 4099007 A1), which evenly introduces the magnetic flux into the tube. On the side of the coil radially distant from the tube, the core is mounted on a ring yoke, which acts as a return conductor for the magnetic flux. The core, ring yoke, and pole piece of the second magnetization device 320 are made of ferromagnetic material in order to offer the lowest possible magnetic resistance to the magnetic flux.

[0047] Longitudinal defects, i.e. defects that run in the z-direction, generate a stray flux signal when they disturb a magnetic flux in the cp-direction. The magnetic flux in the (p-direction is generated by the two oppositely polarized pole pieces. Between the (p- positions of the pole pieces (at 0° and 180°) there is a probe head (at -90° and -270°), each containing a probe array with up to 100 or more magnetic field-sensitive probes. The pole pieces and the probe heads are located in test zone 330 of the second subsystem. So that the entire surface of the pipe can be tested, the entire second magnetization device, i.e. coils, ring yoke and pole pieces including the probe heads and associated electronics, rotates around the pipe while the pipe is transported in the z-direction. The sensors built into the probe heads thus scan the pipe surface in a helical manner.

[0048] It should be noted that while the length of the magnetization vectors is the same at the location of the probes, the direction of the vectors is opposite for each of the two probes. After the pipe has left the test zone of the second subsystem 300, it is demagnetized again by the rotating magnetic field. The demagnetization occurs almost entirely in the cp direction and at least partially in the z direction, since the magnetic field also has a z component at the pole shoe ends.

[0049] To ensure that the pipe can be processed or used without problems after the flux leakage test, standards or the pipe users define the maximum residual magnetization that may remain in the pipe. The residual magnetization in the z-direction in particular is considered critical. A typical procedure for determining the residual magnetization in the pipe is to measure the magnetic flux escaping at the pipe ends using a mobile magnetometer. Depending on the standard, the average value of the magnetic flux over the pipe circumference at one pipe end must not be greater than 3.0 mT, for example. Pipe users sometimes require much lower permissible flux densities, e.g., less than 1 mT. To systematically achieve these low values, partial demagnetization by the rotating second subsystem is often not sufficient.

[0050] In conventional testing devices (sometimes referred to as flux leakage lines), a demagnetizing coil is therefore positioned behind the second subsystem, i.e., on the side facing away from the first subsystem. The demagnetizing coil is usually designed as a simple air-core coil, i.e., it consists of only one or more copper windings without flux conductors or concentrators. The copper windings are operated with an alternating current, the frequency and amplitude of which must be adjusted based on the diameter, wall thickness, and transport speed of the pipe (see, for example, DE 3819033 C2).

[0051] The inventors have recognized that this conventional concept has specific disadvantages. They propose deviations from this traditional concept that avoid these disadvantages and enable improved performance of the test device.

[0052] A first disadvantage of the conventional arrangement is that, depending on the design and operating parameters of the AC demagnetization coil, the effect of the alternating field at high currents may be perceived not only in the adjacent rotating subsystem but also in the first subsystem much further away and may lead to impairments of the test signals.

[0053] A further disadvantage concerns a previously unrecognized adverse effect of demagnetization on the testing accuracy for slant defects. For understanding, the following is explained in connection with Figs. 4A and 4B as well as 5A and 5B. Sub-figures 4A and 5A each show the magnetization situation between pole pieces (field F), while the corresponding sub-figures 4B and 5B each show the associated fault signal amplitudes SIG at the probes PK1 and PK2. A pole piece (letters S and N stand for magnetic south pole and north pole on a pole piece, respectively) generates a magnetic flux in a non-biased tube that is oriented almost exclusively in the (p-) direction in the area of ​​the probes PK1, PK2. A maximum leakage flux signal is generated by a LF that is oriented exactly orthogonal to the cp direction. For slant defects LH and RH, the oblique orientation of the defect leads to a reduction in the leakage flux signal compared to the LF.The reduction of the leakage flux signal in the pipe magnetized exclusively in the (p-direction) is independent of whether the defect is an LH or an RH defect. Since the difference in the directions of the magnetization vectors M- and M2 between the first probe PK1 and the second probe PK2 is 180°, this has no influence on the leakage flux signal because the angles between the magnetization vectors MT and M2 and the defect are the same.

[0054] These are the conditions when the pipe is not magnetized before entering the second subsystem.

[0055] However, in the flux leakage test device considered here, a tube is first tested for transverse defects. Therefore, the tube arrives with a residual magnetization in the z-direction (M z ) in the rotating second subsystem. This residual magnetization M directed in the longitudinal direction of the tube zand the circumferential magnetization of the second subsystem (M <p) addieren sich zu einem resultierenden Magnetisierungsvektor MT bzw. M2. Der Unterschied der Richtungen des Magnetisierungsvektors M zwischen PK1 und PK2 beträgt nach der Addition des Vektors M z no longer 180°, ie the angles at which the two magnetization vectors meet an oblique defect are different.

[0056] If a right-handed skew defect RH is detected with the first probe PK1, the magnetization vector MT is almost perpendicular to the defect, resulting in an almost maximum leakage flux signal. In contrast, with the second probe PK2, the magnetization vector M2 is almost parallel to the defect, resulting in only a very small leakage flux signal. For an oppositely oriented defect LH, the probe signals behave in exactly the opposite direction.

[0057] The z-component of the magnetization in the area of ​​the second subsystem, which is responsible for the differences in the skew error signals, is influenced not only by residual magnetization components after passing through the first subsystem, but possibly also by a demagnetization coil located in the line behind the second subsystem.

[0058] This effect, first recognized by the inventors, has a significant impact on the reproducibility of skew defects. Reproducibility is an important criterion for the quality of flux leakage test instruments. Reproducibility here means that comparable defects should be detected with a comparable signal amplitude, regardless of which of the two probes detects the defect. Test instruments often have to demonstrate during customer acceptance that they can find comparable defects with a reproducibility of better than 3 dB. The difference in the signal strength of an LH or RH skew defect detected by the rotating subsystem can be up to 2 dB or more for a pipe pre-magnetized by the first subsystem.Since, with typical settings, a defect is only fully detected by one of the two probes, this 2 dB of reproducibility for angular defect detection must be added to the normal reproducibility caused by other influences, which is also approximately 2 dB. Therefore, the acceptance criterion may not be systematically met.

[0059] The invention addresses the problems mentioned and provides a novel design of a generic testing device: This is characterized, among other things, by a significantly improved reproducibility of oblique errors and by the fact that the test specimen only has residual magnetization of uncritical strength after passing through the testing device.

[0060] One aspect of the proposed solution to the problem is that the magnetization component M zshould be eliminated as completely as possible before the corresponding pipe section is tested by the second subsystem.

[0061] For this purpose, a special arrangement of a demagnetization coil 400 is selected with which it can be achieved that the residual magnetization remaining in the pipe after passing through the test zone of the first subsystem 200 is reduced and minimized as far as possible by a DC field magnetization with opposite polarity before the pipe enters the second subsystem (see Fig. 1). The outlet side of the first subsystem 200 has proven to be the ideal location for the demagnetization coil 400 acting as a compensation coil.

[0062] The demagnetization coil 400, which can also be referred to as a compensation coil, can be structurally identical to the two magnetization coils 222-1, 222-2 of the first magnetization device 220. In the example, the demagnetization coil is mounted directly on the housing of the first subsystem 200 on its outlet side, i.e., without any clearance. The distance to the second subsystem is many times greater than any possible distance to the first subsystem.

[0063] Due to the spatial proximity (very small air gaps, thus low magnetic resistance) of the demagnetization coil 400 to the flux guides present in the first subsystem, a smaller coil (smaller winding space) and / or a lower electrical power can be used. The local effect of the longitudinal field demagnetization coil on the pipe to be demagnetized can be precisely controlled, so that the remaining residual field in the pipe can be minimized. The demagnetization coil 400 is operated separately with a current-controlled DC power supply (adjustable DC source 424).

[0064] Immediately after the section to be tested has left the first subsystem, the demagnetization coil 400 (compensation coil, counter coil) acts by building up a constant magnetic field in the -z direction, thus reducing the remanence in the pipe to zero, ideally upon reaching the coercive field strength. If the coercive field strength is not fully reached by the counter coil, the second subsystem 300 also demagnetizes a small portion of the M component with its rotating magnetic field. z , because at the edge of the pole shoes the magnetic field is not exclusively oriented in the cp direction (so-called bulging of the magnetic field).

[0065] If the residual magnetization in the z-direction M z is eliminated, only the magnetization component M <p und das Rohr verhält sich bei der Längsfehlerprüfung, d.h. bei der Prüfung mittels des zweiten Teilsystems, wie ein nicht vormagnetisiertes Neurohr.

[0066] To the same extent as the current intensity of the magnetizing coils of the first subsystem depends on the pipe diameter, wall thickness and the protective nozzles used to achieve a suitable magnetic operating point, the setting of the current of the counter coil l g affected by these dependencies.

[0067] Due to the rather soft magnetic properties of the ferromagnetic tubes, i.e. the coercive field strength is achieved at a smaller magnetic field compared to magnetization, the counter coil usually requires a lower field strength than the magnetization coils 222-1, 222-2.

[0068] In addition to improved skew error reproducibility, a correctly adjusted demagnetization coil offers another advantage: the pipe exhibits almost no residual magnetization after the second subsystem. In most cases, a residual magnetization measurement as described above should yield a value in the range of 0.5 mT. Only for manufacturers or pipe users with the highest demands would an additional demagnetization coil be necessary. However, this demagnetization coil should not be located in close proximity to the two subsystems, but rather at a greater distance, e.g., several meters (e.g., at least 5 m) from the second subsystem, to prevent any influence on skew error reproducibility.Pure DC demagnetization can be used particularly effectively, especially if at least a qualitative measurement of the magnetization state of the material and a precise control of the demagnetization coil are realized.

[0069] During direct current demagnetization, the magnetic domains are oriented in such a way that no magnetic field can be measured outside the material. However, the domains created during direct current demagnetization appear to be larger than those created during alternating current demagnetization, which is why a small amount of residual magnetism can develop again after a certain time or after mechanical action. However, this effect is not observed in the leakage flux test proposed here, which consists of a series connection of longitudinal flaw detection, a direct field demagnetization coil (counter coil), and transverse flaw detection. It is currently assumed that this advantage arises because the 400-large domains created after passing through the direct field counter coil are sufficiently reduced in size by the demagnetizing properties of the downstream rotating subsystem (z-component at the edge of the pole shoes and rotation at, for example, 1 - 20 Hz).

[0070] As mentioned above, a slant flaw calibration procedure can be used to determine the demagnetizing field strength required to sufficiently reduce the longitudinal residual magnetization in the pipe prior to testing for longitudinal flaws. The average value of the calibration of the two probes can be compared. If the flaw signal amplitudes differ, the longitudinal magnetizing field strength is not zero, and the DC current in the demagnetizing coil must be adjusted accordingly.

[0071] In a state-of-the-art comparison, a longitudinal defect is moved at a low feed rate through the measuring range of the probes of the second subsystem, so that each probe of the two probes records the longitudinal defect multiple times. With homogenized pole pieces and ideal probes, the longitudinal defect generates the same maximum signal in each probe, regardless of the probe and the residual magnetization Mz (Figs. 4a, 4b).

[0072] During skew error calibration, the procedure described above is performed on a skew error. This results in a difference in the maximum probe signals between the two probes in the event of residual magnetization Mz being present (Figs. 5A, 5B). Instead of the individual probe signals, the average value of the signals for each probe can also be used. By adjusting the DC current in the degaussing coil, the difference in the signals between the two probes can be reduced; a skew error calibration procedure should be performed for each adjustment. The procedure is successfully completed when the skew error signals for both probes have almost the same amplitude, because the residual magnetization Mz is then also almost zero.

[0073] Some embodiments provide for fully automated adjustment of the strength of the demagnetizing field. For this purpose, a test tube is used in which slant defects of defined size and orientation are introduced at a sufficiently large distance from one another. The spacing of the slant defects should be large enough to allow the DC current strength of the demagnetizing coil to be varied between the defects during slant defect calibration. For example, the demagnetizing current can be high at the first slant defect and comparatively lower at the second slant defect. Depending on the demagnetizing current, two error signal values ​​are then obtained for each test head, for which a best-fit line can be calculated as a function of the demagnetizing current.The two straight lines intersect at a point where the error signals from the two probes coincide and the longitudinal magnetization component Mz disappears. The coil current corresponding to this value then corresponds to the coil current of the demagnetization coil with which the magnetization acting in the z-direction can be reduced to zero. The accuracy of the automated adjustment can be increased by using more than two pairs of oblique error demagnetization current values, so that a higher-order function can be used instead of a best-fit line.

Claims

Patent claims 1. Test method for stray flux testing of ferromagnetic test material, in particular ferromagnetic tubes, for the detection of defects, in a continuous process, wherein the test material is conveyed along a flow direction first through a first subsystem and then through a second subsystem of a test device and is demagnetized before exiting the test device, wherein the first subsystem is designed as a transverse defect testing system and comprises a first magnetizing device which generates a magnetization in the test material whose field lines run essentially in the longitudinal direction of the test material, the second subsystem is designed as a longitudinal defect testing system and comprises a second magnetizing device which generates a magnetization in the test material whose field lines run essentially in the circumferential direction of the test material.In each of the subsystems, a surface of the test specimen is scanned by means of at least one magnetic field-sensitive probe to detect magnetic stray fields caused by defects in the area of ​​a test zone, and electrical probe signals are evaluated to qualify the defects, characterized in that, after passing through the test zone of the first subsystem, the test specimen is largely or completely demagnetized upon exiting the first subsystem and before entering the second subsystem by means of a constant magnetic field of a demagnetizing coil assigned to the first subsystem, which opposes the magnetic field of the first magnetizing device.

2. Test method according to claim 1, characterized in that magnetically conductive components of the first subsystem are also used to conduct the demagnetizing field.

3. Test method according to claim 1 or 2, characterized in that the strength of the DC field of the demagnetizing coil is adjusted to the respective requirement by adjusting a connected DC current source.

4. Test method according to one of the preceding claims, characterized by a residual magnetization determination operation, wherein a residual magnetization value is determined on the test specimen after passing through the first subsystem and the demagnetization coil, which represents the strength of a magnetization component directed in the longitudinal direction of the test specimen. represented, and by regulating the strength of the constant magnetic field generated by the demagnetizing coil as a function of the residual magnetization value.

5. Test method according to claim 4, characterized in that in the residual magnetization determination operation the second subsystem is used as a measuring system to determine the residual magnetization value.

6. Test method according to one of the preceding claims, characterized in that the second subsystem has a rotating head which has pole shoes of the second magnetizing device at diametrically opposite locations radially aligned to the surface of the test specimen, such that a constant magnetic field can be generated inside the test specimen, the field lines of which run in the circumferential direction of the test specimen, that at least one probe is arranged in the circumferential direction between the pole shoes, and that a skew error adjustment procedure is carried out, wherein the strength of the constant magnetic field generated by the demagnetizing coil is adjusted such that a skew error generates substantially the same error signal amplitude in each probe scanning the skew error.

7. Test device (100) for leakage flux testing of ferromagnetic test material (110), in particular ferromagnetic tubes, for the detection of defects, in a continuous process, comprising: Conveying devices for conveying the test item along a through-direction (102) through a first subsystem (200) and a second subsystem (300) of the test device downstream of the first subsystem in the through-direction, wherein the first subsystem (200) is designed as a transverse defect testing system and has a first magnetizing device (220) configured to generate magnetization in the test item whose field lines run substantially in the longitudinal direction of the test item, the second subsystem (300) is designed as a longitudinal defect testing system and has a second magnetizing device (330) configured to generate magnetization in the test item whose field lines run substantially in the circumferential direction of the test item, and each of the subsystems has a test zone (230, 330) with at least one magnetic field-sensitive probe (250, 350),with which a surface of the test item can be scanned to detect magnetic stray fields caused by defects; an evaluation unit (500) for evaluating electrical probe signals of the probes to qualify the defects, A demagnetizing device designed such that the test specimen can be demagnetized to a tolerable residual magnetization before exiting the test device by means of the demagnetizing device, characterized in that the demagnetizing device has a direct current source (424) and a demagnetizing coil (400) connected thereto, which is arranged in an outlet area of ​​the first subsystem (200) and can be operated with direct current such that the test specimen magnetized by the first magnetizing device (220) can be demagnetized by reducing the magnetization component running in the longitudinal direction of the test specimen.

8. Test device according to claim 7, characterized in that the demagnetizing coil (400) is attached to a housing of the first subsystem (200).

9. Test device according to claim 7 or 8, characterized in that the first magnetizing device (220) has a magnetizing coil (222-1, 222-2) and test specimen passage nozzles (225-1, 225-2) made of ferromagnetic material on an inlet side and an outlet side, which are coupled via magnetic return conductors (227), and that the demagnetizing coil (400) is arranged on the first subsystem (200) such that field lines of a demagnetizing field generated by the demagnetizing coil pass through the flux conductors on the outlet side of the first subsystem (200).

10. Test device according to claim 7, 8 or 9, characterized in that the DC current source (424) of the demagnetizing device is designed as an adjustable DC current source, so that the strength of the demagnetizing field is adjustable.

11. Test device according to one of claims 7 to 10, characterized by a control loop for automated control of the strength of the constant magnetic field generated by the demagnetizing coil (400) as a function of a residual magnetization value that can be determined in a residual magnetization determination operation.

12. Test device according to claim 11, characterized in that the second subsystem (300) is configured as a measuring system for determining the residual magnetization value such that the second subsystem can be used as a functional part of the control loop for setting the correct demagnetization field strength.

13. Test device according to one of claims 7 to 12, characterized in that the second subsystem (300) has a rotating head which has pole shoes (325-1, 325-2) of the second magnetizing device (320) at diametrically opposite locations radially aligned to the surface of the test specimen, such that a constant magnetic field can be generated inside the test specimen, the field lines of which run in the circumferential direction of the test specimen, wherein at least one probe (350) is arranged in the circumferential direction between each of the pole shoes, wherein the device is configured to carry out a skew error adjustment procedure, wherein the strength of the constant magnetic field generated by the demagnetizing coil (400) is adjusted such that a skew error (SF) generates essentially the same error signal amplitude in each probe scanning the skew error, regardless of its skewness.

Images