DE102014209602A1 - Test method and test device for non-destructive material testing by means of a probe array - Google Patents

Abstract

In a non-destructive testing of a device under test 190, a probe array 110 is used which has a plurality of probes P1, P2, ... Pn. The probe array is positioned near a surface 192 of the device under test. Probe signals from a plurality of probe probes of the probe array are added to a sum signal by means of a signal adding device ADD in an addition operation, and the sum signal resulting from the addition operation is evaluated to characterize defects in the device under test in an evaluation operation. The method is characterized in that the probe signals before the addition operation by means of an amplitude filter AF are subjected to amplitude filtering with respect to a predetermined threshold, wherein probe signals with a signal amplitude below the threshold filtered out and only probe signals with a signal amplitude greater than or equal to the threshold value of the addition operation become.

Description

AREA OF APPLICATION AND PRIOR ART

The invention relates to a test method for non-destructive material testing of a test specimen by means of a probe array comprising a plurality of test probes, according to the preamble of claim 1 and a test apparatus suitable for carrying out the test method according to the preamble of claim 6.

In non-destructive testing (NDT), for example in the automated non-destructive testing of semi-finished products for the metal-producing and metalworking industry, to carry out tests on safety-critical and mission-critical components for land vehicles and aircraft or in plant construction, nowadays come after different Principles working test methods and corresponding test devices used. For example, the eddy current testing has proven in many applications in the testing of electrically conductive materials. The test probes used for eddy current testing are commonly referred to as "eddy current probes". Stray flux testing, ultrasonic testing, GMR (Giant Magneto-Resistance) probing or other magnetic field sensitive probes are also commonly used.

The small-defect sensitivity that can be achieved with a test method is usually determined by the size of the sensitive area of a test probe, so that probes for defect detection usually have small sizes. On the other hand, it is often desired to test larger surface areas of a test specimen in a relatively short time. Frequently, so-called probe arrays are used here, which have a large number of (normally identical) test probes in a predetermined spatial arrangement.

In some probe arrays with two or more probes, the probe signals of each of the probes can be evaluated independently of the probe signals of the other receiver coils. It is also possible to provide a common evaluation of the probe signals from some or all probes. Among other things, there are test devices in which signal outputs of the probes are electrically connected in series or can be switched, so that the probe array produces an overall sum signal, which results from the addition of voltages or voltage components generated in the individual probes , This will make it u.a. possible to check larger areas with higher error resolution.

If test probes connected in series are used, it may be possible to dispense with complex multiplex technology since probe signals of all test probes connected in series can be evaluated via a single evaluation channel.

As a rule, small defects result in probe signals with a relatively small signal amplitude. Small defects can be functionally critical for the examinee. Test probes should therefore have high sensitivity even for small defects. However, if functionally uncritical irregularities are also identified as potential defects, there is a risk of discarding defect-free specimens or test specimen sections as being defective, as a result of which unnecessarily much waste may be produced. Test methods and test equipment should be designed to allow reliable detection of defects and reliable discrimination between sought-after defects and irrelevant pseudo-defects.

TASK AND SOLUTION

It is an object of the invention to provide a generic test method and a generic test apparatus, which allow reliable detection of relatively small defects in materials with many small irregularities, if required. In particular, a reduction of the signal-to-noise ratio should be achieved compared to conventional solutions.

To achieve these and other objects, the invention provides a test method having the features of claim 1 and a test apparatus having the features of claim 6. Advantageous developments are specified in the dependent claims. The wording of all claims is incorporated herein by reference.

In the test procedure, the probe array is positioned near a surface of the device under test so that the probes can interact with the device under test. Often a small distance (test distance) remains between the surface of the device under test and the test probe. However, this is not mandatory, there are also variants with test probes, which are brought into direct contact with the sample surface for testing. In carrying out the test, probe signals from several probes of the probe array are then added in an addition operation to a sum signal. It is possible to form the sum signal only from the probe signals of a subgroup of all test probes of the probe array, or from the probe signals of all probes of the probe array. The summation signal resulting from the addition operation is then used to characterize defects in the examinee evaluated in an evaluation.

In the claimed invention, it is now provided that the probe signals are subjected to amplitude filtering with respect to a predetermined threshold prior to the addition operation, with probe signals having a signal amplitude below the threshold being filtered out and only probe signals having a signal amplitude greater than or equal to the threshold value of the addition operation ,

In a correspondingly configured test device, an amplitude filter device is electrically connected between the test probes and the signal adding device, which effects this amplitude filtering.

The claimed invention is based i.a. Based on the knowledge that in generic test methods and test devices when using probe arrays may result from irregularities in the test specimen resulting noise background can be so strong that the resolution of the test system is reduced for small defects. The claimed invention remedy this by performing amplitude filtering prior to the addition operation. This ensures that in the addition operation no noise signal components are added that come from areas in which the signal amplitude is below the threshold. If the threshold value is correctly designed, the probability is high that such areas are free from the sought-after, possibly critical defects. As a result, the signal-to-noise ratio (S / N ratio) is improved as compared with conventional solutions.

The amplitude filtering upstream of the addition operation can have a favorable effect on the detection of any type of defects. Particular advantages are currently seen in the detection of relatively small defects whose signal amplitude is only slightly above the threshold. These can be reliably detected, since their signal amplitudes completely pass through the amplitude filtering and can reach the signal adding device. Thus, an improvement in the signal-to-noise ratio can be achieved for single-channel evaluated signals from probe arrays.

The probe array may be stationary or fixed to a component to be monitored. In other embodiments, the device under test and the probe array are moved relative to one another along a direction of movement. Such a relative movement makes it possible to test a wide test track on the specimen surface with high spatial resolution, the width of the test track corresponding to the width of the probe array perpendicular to the direction of movement) and the spatial resolution u.a. is determined by the extent of the individual test probes. Probe arrays can be used in scanning systems, e.g. for a run test. At least one probe array may e.g. be installed in a rotating head of a tester to rotate in a run test around a DUT and to scan the DUT surface along a helical path.

In order to allow a reliable distinction between any critical defects and non-critical artifacts that appear to be defective, in a preferred embodiment a calibration operation is performed, the threshold depending on the type of specimen and / or other criteria (eg shape of the specimen, material properties of the specimen etc.) is set. In this way, the discrimination of probe signals can be individually adapted to a candidate or to a class of DUTs.

In a variant of the method, the average noise signal amplitude of a noise signal is determined during the calibration operation on a defect-free reference test specimen of the test specimen in question or a defect-free specimen section of a specimen to be tested, preferably in conjunction with a standard deviation or another parameter representing a fluctuation width of the signal amplitude. The threshold value is then set at a predetermined distance above the noise signal amplitude, so that even individual noise signals that slightly exceed the average noise signal amplitude, in case of doubt, will be filtered out. For many cases, it has proven to be expedient if the predetermined distance is between approximately 10% and approximately 20% of the mean noise signal amplitude. If the lower limit is exceeded, the risk increases that unnecessarily frequent noise signals from defect-free areas reach the summation and thus the overall signal is noisy. If, on the other hand, the upper limit is exceeded, the risk increases that small but critical effects are not recognized as such because the associated probe signals are filtered out before the addition operation.

The invention also relates to a test apparatus configured for carrying out the test method for the non-destructive testing of a test specimen.

The probes may e.g. be designed as eddy current probes, as stray flux probes or as GMR probes, so that magnetic methods can be used for testing.

In one embodiment, the probe array comprises a plurality of eddy current probes, which has an exciter arrangement with at least one electrical conductor for connection to an AC voltage source and a receiver arrangement with at least one receiver coil with one or more windings separate from the exciter arrangement. The excitation arrangement has an exciter section in which a conductor section or a plurality of conductor sections run in such a way that a primary alternating magnetic field with magnetic field lines extending around the excitation section is generated by the conductor section or the conductor sections of the exciter section when connected to an AC voltage source. The receiver coil is arranged with respect to the exciter section such that the receiver coil is symmetrically penetrated by the primary alternating magnetic field of the exciter section such that a time change dφ / dt of the magnetic flux φ of the primary alternating magnetic field by the receiver coil substantially disappears. As a result, a magnetic flux compensation of the primary field in the receiver coil can be achieved. Defects in the specimen disturb this symmetry and then lead to measuring voltages that can be evaluated. Since theoretically no measuring voltage is generated in a defect-free reference situation, the evaluation can be carried out with high sensitivity.

Preferably, several or all of the eddy current probes of the probe array have a common excitation arrangement. As a result, the test probes have no differences due to different excitation, whereby more precisely interpretable test results can be achieved.

BRIEF DESCRIPTION OF THE DRAWINGS

Further advantages and aspects of the invention will become apparent from the claims and from the following description of preferred embodiments of the invention, which are explained below with reference to the figures. Showing:

1 shows a schematic plan view of a test head of a tester for non-destructive testing of a specimen;

2 shows schematically a test device according to the prior art;

3 shows schematically a equipped with an amplitude filter device testing device according to an embodiment of the invention;

4 shows an embodiment with an analog amplitude filter device;

5 shows an embodiment with a digital amplitude filter means;

6 shows in plan view an eddy current testing device with a probe array according to one embodiment of the invention; and

7 shows a vertical section through the probe array along the line VII in FIG 6 ,

DETAILED DESCRIPTION OF THE EMBODIMENTS

Based on 1 to 3 First, some basic principles of construction and operation of inventive test devices are illustrated using an exemplary embodiment. 1 shows a schematic plan view of a test head 100 a test device for non-destructive testing of a DUT 190 , which consists in the example of a metallic material. The test head is used to perform a test at a small distance (test distance) to the surface to be tested 192 brought the test specimen. The test head is mounted stationary in the example, while the test specimen to carry out the test along a direction of movement 198 is moved relative to the test head.

The specimen has some roughness on its surface and some smaller pores and other irregularities 194 on, which are not critical for the function of the specimen when used as intended. Furthermore, a defect 196 to recognize in the form of a tear reaching to the surface, which could serve as the starting point of a crack under mechanical stress of the specimen in the subsequent use and therefore is a function critical defect. By means of the test apparatus such function-critical defects should be detected reliably, at the same time a reliable distinction between functionally critical defects and non-critical irregularities should be achieved.

In the test head 100 there is a probe array 110 comprising a plurality of individual probes P1, P2, ..., Pi, ..., Pn. The probes are arranged in a single rectilinear series with small, regular mutual distances to each other, thereby forming a linear probe array. The mutual distance of immediately adjacent test probes is dimensioned such that the test areas or action widths covered by the individual test probes overlap each other slightly, so that the row of test probes covers a test width PB determined by the extent of the probe array without gap. In other embodiments, a probe array has two or more parallel rows whose Test probes laterally offset from each other are arranged so that they are arranged in the direction of movement "to gap" to each other and thus allow for relatively small sensitive areas of the individual probes a complete test over a test width.

In this application, the term probe array primarily describes a multi-element sensor with many probes, in which the relative position of individual probes is periodic (e.g., line or matrix). However, depending on the application or the test task, an irregular arrangement of test probes is also possible.

The test probes are signal-conducting, for example wirelessly or via electrical lines, with a common evaluation device 120 the testing device connected. During the execution of a test, each of the test probes generates its own probe signal in the form of an electrical voltage whose time profile, amplitude (signal amplitude) and possibly phase position relative to a reference voltage is used for the evaluation.

In the test apparatus comprising the probe array, it is provided that the signal information of all test probes of the probe array is to be supplied to the evaluation unit EV via a single evaluation channel. For this purpose, the probe signals of several or all probes of the probe array are added by means of a signal adding device ADD in an addition operation to a sum signal, which is then evaluated to characterize defects in the DUT in an evaluation operation. On the one hand, each test probe detects signal components that are sensitive to (uncritical) irregularities 194 in the overrun area, and on the other hand signal components, which are due to any critical defects. In the example case, only the second test probe P2 will generate a defect signal component if the second test probe reports the defect 196 away travels. The other probes generate only on the irregularities 194 decreasing noise in the probe signal. Prior to the adding operation, an amplitude filtering of the probe signals is performed by means of an amplitude filter AF.

The 2 and 3 show possible Verschaltungsanordnungen for implementing a test method in which the test signals of several probes of a probe array are first added in a signal adding device ADD and the resulting sum signal SUM is then evaluated in an evaluation EV. It shows 2 an arrangement according to the prior art and 3 a possible embodiment of an embodiment of the invention.

In the example of the prior art 2 the test signals of the probes P1 to Pi arrive unfiltered to the signal adding device ADD, which generates the sum signal SUM, which is subsequently evaluated in the evaluation unit EV. Starting from the in 1 The resulting sum signal SUM contains the sum of all noise signal components detected by all test probes, which are based on the uncritical irregularities 194 go back, and, when driving over the defect 196 through the second test probe P2, in addition to the signal component due to this defect. Since all of the noise signal components are added in the addition operation, the total noise signal amplitude resulting from the summation may be of the order of magnitude due to the defect 196 caused defect signal amplitude, so that the signal / noise ratio can be unfavorably small. This is the reason why the use of signal addition probe arrays in some test environments (eg with small imperfections) may affect the resolution of the test procedure.

Such problems can be avoided or reduced in embodiments of the claimed invention. In the embodiment in 3 is an amplitude filter AF electrically connected between the probes P1 to Pi of the probe array and the signal adding device ADD. The amplitude filter AF is designed so that each of the probe signals are each compared with a predetermined threshold value of the signal amplitude. In this case, probe signals which have a signal amplitude below the threshold value are completely filtered out, while probe signals with a signal amplitude which corresponds to the threshold value or is greater than the threshold value are completely transmitted (including noise component) to the signal addition device ADD.

The predetermined threshold is set in the example case, that signal components resulting from the irregularities 196 of the DUT, are likely to be below the threshold, while signal amplitudes resulting from critical defects are in any case above the threshold. In the test situation off 1 follows from the fact that the probe signals of all test probes, which run over only the areas of uncritical irregularities when the test specimen is moved away, are filtered out by the amplitude filtering and do not contribute to the summation. By contrast, the test signal of the second test probe P2 has due to the defect 196 caused signal component has a signal amplitude which is above the threshold value and is therefore transmitted including its noise components to the signal adding device ADD and in the Addition operation considered. The resulting sum signal SUM thus contains only that noise component which is detected by the second test probe P2 and that due to the defect 196 caused, relative to this much larger defect signal component. This is clearly distinguished from the noise background, because the amplitude filtering causes the noise signal component to be compared with an addition without prior amplitude filtering (FIG. 2 ) is substantially reduced. In other words: by the amplitude filtering before the signal addition, the signal / noise ratio compared to a corresponding evaluation without amplitude filtering can be significantly improved.

The magnitude of the threshold, which differentiates between unwanted noise signal components and sought defect signal components, may be preset based on previous experience over type-identical or type-like samples. In some embodiments, the threshold value on the testing device is infinitely or in stages adjustable, so that the test device can be optimally adjusted by adjusting the threshold value to the respective test situation and the type of the test object.

The threshold value can be set by means of a calibration operation. In a case with an ideal test probe and amplitude analysis, a certain voltage of the noise signal is measured and then used as a threshold value. The threshold may be adjusted depending on the type of device under test and / or on the type of probe (e.g., absolute probe or differential probe).

In the calibration operation, e.g. On a defect-free reference test piece or a defect-free Prüflingsabschnitt a noise signal amplitude of a noise signal can be determined and the threshold can then be set at a predetermined distance above the noise signal amplitude. In many cases, it has proven to be favorable if the predetermined distance is between 10% and 20% of the noise signal amplitude.

The amplitude filter device can be realized analog or digital depending on the application. In 4 an embodiment is shown in which the test signals of the probes P1 to Pi an analog amplitude filter AF-AN are supplied before the non-filtered test signals are added using a signal adding device ADD and then evaluated in the evaluation EV. The amplitude filter device comprises for each test channel, ie for each test probe, a voltage comparator V and a controllable by the voltage comparator switch S, which can be switched between a closed state (signal passage) and an open state (filtering the signal). The comparator V compares the signal amplitude of the probe signal or the corresponding signal voltage USig with a threshold voltage corresponding to the threshold UT (Threshold Voltage). If the signal voltage is at least equal to the threshold voltage or greater (USig ≥ UT), the switch is closed, so that the probe signal is supplied as a whole of the addition and the evaluation. At signal voltages below the threshold voltage (USig <UT), the switch is opened, so that the probe signal is filtered out as a whole and thus does not contribute to the formation of the sum signal.

In 5 an example of a digital realization of the amplitude filtering or the amplitude filter device is shown. The digital amplitude filter device AF-DIG comprises an analog / digital converter A / D for each test channel or for each test probe, which generates a digital signal from the analog test probe signal. The digital signals of the individual test channels are applied to the inputs of a programmable integrated circuit in the form of a Field Programmable Gate Array (FPGA). In this component, a logic circuit is programmed, which compares the digital input signals in each case with a digital signal for the set threshold value. Only those digital signals corresponding to probe signals whose signal amplitude is above or equal to the threshold value are applied to the integrated adding operation and contribute to the sum signal SUM. The other signals whose signal amplitude is below the threshold are filtered out and do not reach the evaluation device EV.

In the embodiments, the threshold is the same for each test channel. However, this is not mandatory. Variants are also possible which realize different threshold values for different test probes. This can be useful, for example, if the test object consists of several materials and / or has changes in shape in the tested section.

The basic principles of the claimed invention may be utilized in different types of test devices and test methods. For example, the probes could be so-called GMR probes that use the GMR effect (Giant Magnetoresistance) for test signal generation. It could also be stray flux probes, eddy current probes or other magnetic probes. Test probes for imaging techniques (eg, magneto-optic imaging or optical measurements, combinations of eddy current and laser measurements, etc.) may also benefit from the use of the claimed invention. Tester) with probes in the form of eddy current probes explained in more detail.

In the eddy current test, an eddy current probe is arranged at a small distance (test distance) to the surface of the electrically conductive test specimen. An eddy current probe typically has a excitation arrangement with at least one electrical conductor which is provided for connection to an AC voltage source, and a receiver arrangement with at least one receiver coil which has one or more windings and is often referred to as a measuring coil. Eddy current probes in which the exciter coil and the receiver coil are formed by the same coil are referred to as parametric eddy current probes. When an eddy current probe has an exciter coil and a receiver coil separate therefrom, with coupling between the exciter coil and the receiver coil being mediated via the specimen material, one usually speaks of transformer eddy current probes. Within the scope of the claimed invention, all types of eddy current probes may be used, with transformatory ones being preferred.

The excitation coil is connected to an AC voltage source to perform the test and can then generate a primary electromagnetic alternating field (primary magnetic field) which penetrates the test material during testing and generates eddy currents substantially in a near-surface layer of the test material by mutual induction on the receiver coil (n) react back the eddy current probe. The secondary magnetic field (secondary magnetic field) caused by the eddy currents is opposite to the primary magnetic field according to Lenz's rule. A defect in the tested area, such as a crack, contamination or other material inhomogeneity, interferes with the propagation of the eddy currents in the test material and thus alters the eddy current intensity and thereby also the intensity of the secondary magnetic field applied to the receiver coil. The resulting changes in the electrical characteristics of a receiver coil, e.g. in particular the impedance of the receiver coil lead to electrical measurement signals that can be evaluated by means of an evaluation device to identify and characterize defects.

Based on 6 and 7 is an embodiment of an eddy current testing device 600 explained in more detail. The test head of the test device has a pathogen arrangement 610 on, the four identical, arranged in a straight line receiver coils 640A to 640D assigned. Each of the receiver coils forms together with the associated part of the excitation arrangement 610 an own eddy current probe, so that a probe array is present with four individual probes. 6 shows a plan view, 7 a vertical section through the probe array in the region of a receiver coil 640 along the line VII in 6 ,

The exciter arrangement 610 has two excitation coils 610-1 and 610-2 , with their facing inner conductor sections 610-1A respectively. 610-2A a rectilinear exciter section 615 form with two mutually parallel conductor sections, which when connected to an AC voltage source 620 To be traversed in phase in the same direction by the excitation current. The four receiver coils 640A to 640D are arranged in a straight receiver coil series symmetrical to the excitation section so that in relation to the primary alternating field generated by the exciting section in each of the measuring coils in air later explained a magnetic flux compensation and thus in principle results in a voltage of 0 volts as a measurement signal.

The probe array is fabricated using printed circuit board technology. The receiver coils 640A - 640D are each flat coils whose turns in a common plane (figure plane in 6 ) and a receiver coil axis 644 define that is perpendicular to the receiver coil plane. The windings of the receiver coils are on a surface of an electrically insulating carrier layer 670 applied. The electrical conductors of the exciter arrangement are also on a surface of an electrically insulating carrier layer (same carrier layer 670 or another carrier layer) in such a way that at least one electrically insulating carrier layer lies between these electrical conductors and the windings of the receiver coil, so that they are electrically insulated from one another.

The electrical conductors of the excitation system 610 form a first exciter coil 610-1 and a second excitation coil 610-2 , which may each have one or more turns and are each formed as flat coils with coil planes parallel to the receiver coil plane. The turns of the first and the second excitation coil are mirror-symmetrical to a plane of symmetry 645 arranged centrally through the receiver coil and the receiver coil axis 644 contains.

The first exciter coil 610-1 has one of the second exciter coil 610-2 facing straight inner conductor section 610-1A and an outer conductor portion spaced from the second exciting coil at a greater distance from the plane of symmetry 610-1B , which is parallel to the inner conductor section 610-1A runs. Accordingly, the second exciter coil 610-2 a straight inner conductor section 610-2A near the first one Excitation coil, as well as a remote from the first excitation coil straight outer conductor section 610-2B , which is parallel to the inner conductor section 610-2A runs.

When connected to the AC voltage source 620 the two become symmetrical to the symmetry plane 645 arranged, located close to each other and centrally across the receiver coil away extending inner conductor sections 610-1A and 610-2B in the same direction of the excitation current, while the outer outer conductor sections 610-1B and 610-2B each in opposite directions to the inner conductor sections, but in the same direction with each other and that is, in the same direction to be traversed by the current. The rectangular excitation coils are thus mirror-symmetrical to the plane of symmetry 645 and are also electrically operated mirror-symmetrically to this plane of symmetry.

The inner conductor sections 610-1A and 610-2A together form the rectilinear exciter section 615 the exciter arrangement. The pathogen section 615 runs perpendicular to the receiver coil axis in the middle diagonally across the receiver coil 640 , The inner conductor sections become when connected to the AC voltage source 620 go through the same phase of alternating current and produce together a primary alternating magnetic field PF, whose field lines FL around the exciter section 615 to run around (cf. 7 ).

The receiver coil 640 is in relation to the pathogen section 615 arranged in the manner "magnetically symmetrical" that the receiver coil is indeed traversed by the primary alternating magnetic field PF of the excitation section, but the receiver coil is so symmetrically interspersed by the primary alternating magnetic field that a temporal change dΦ / dt of the magnetic flux Φ of the primary magnetic Alternating field by the receiver coil total possible completely disappears. This has the consequence that although the primary magnetic field penetrates the receiver coil, it does not induce any electrical voltage in this receiver coil. This effect is referred to herein as "magnetic flux compensation" or "primary field compensation". This is achieved by the symmetry plane 645 the receiver coil 640 such is that the receiver coil in two magnetically equivalent sub-areas 646-1 and 646-2 is divided, wherein the magnetic flux change through one of the faces is equal to the same flux change through the other face, so that compensate for the changes in the magnetic fluxes. When Φ _{1 is} the magnetic flux through the first face 646-1 and Φ _{2 is} the magnetic flux through the second subarea 646-2 is, then applies to the induced voltage V, the condition: V = - (dΦ ₁ + dΦ ₂ ) / dt ≈ 0

If the excitation section is traversed by alternating current, a primary magnetic field PF is generated whose field lines FL run in planes lying perpendicular to the path of the exciter section around the exciter section. A substantial portion of the magnetic field lines in the vicinity of the excitation section passes through the receiver coil and penetrates into near-surface regions of the test specimen, where eddy currents EC are induced.

A special feature is that with this arrangement in the receiver coil 640 no voltage is induced overall, although an alternating magnetic field passes through the coil. The receiver coil 640 Although an absolute coil in the sense that no signal compensation with another receiver coil. Nevertheless, in this arrangement, the receiver coil also has characteristics of a differential coil arrangement. In this arrangement, namely, a compensation of the induced voltages is achieved in that the receiver coil is permeated magnetically symmetrically by the primary alternating magnetic field of the excitation section in such a way that the temporal change dΦ / dt of the magnetic flux of the primary alternating magnetic field by the receiver coil altogether disappears.

In this case, each component of the magnetic flux which does not run parallel to the receiver coil plane generates an induced voltage as it passes through the coil plane. Due to the magnetic symmetry, however, there is a corresponding oppositely directed component, which would induce a corresponding voltage with opposite polarity, so that overall no output signal (measurement voltage = 0 V) results.

The signal compensation thus takes place by direct compensation of magnetic field fluxes of the primary field in the receiver coil. This magnetic field compensation or primary field compensation means that when this eddy current probe is placed in or near a defect-free material, the total flux changes of the primary alternating magnetic field generated by the exciting section and the secondary magnetic field generated by the eddy currents of the device under test disappear altogether, so that in the receiver coil in FIG In this case no voltage is induced. The between the outputs of the receiver coil 640 Measured electrical voltage is therefore equal to 0 in this case. It is important to note that this magnetic flux compensation is required only for the area within the receiver coil. By contrast, the conditions outside the detection range of the receiver coil have no influence on the output signal of the receiver coil.

The effect of the primary alternating field is better spatially concentrated or focused in this arrangement than in the case of conventional eddy current probes. An improved magnetic field concentration in the central region of the test volume centrally under the receiver coil can be assisted by having on both sides next to the exciter section 615 conductor sections 610-1B and 610-2B the exciter arrangement are provided, which are also traversed by the excitation current, but in the opposite direction to the exciter section. As a result, alternating magnetic fields are generated on both sides of the excitation section, which are directed counter to the alternating magnetic field of the exciter section. These alternating fields weaken the primary alternating field of the pathogen section 615 outside the test area, so that the primary alternating field is concentrated or focused more in the test volume below the pathogen section. "Focusing" here means that the spatial density of the magnetic field lines in the focus is increased. Due to this local concentration of the effective range of the primary magnetic field PF on the area centrally below the receiver coil 640 a particularly efficient defect inspection with high spatial resolution is possible.

Each of the receiver coils 640A to 640D is connected to a channel on the input side of an amplitude filter device AF, which makes it possible to record the output signals of the individual receiver coils (ie the probe signals) in separate channels and to filter them by evaluating their signal amplitude with respect to a threshold value. The amplitude filtering takes place in such a way that probe signals with a signal amplitude below the threshold are filtered out and only probe signals with a signal amplitude greater than or equal to the threshold value of the addition operation are supplied. The four output channels of the amplitude filter device AF are connected to separate input channels of a signal adding device ADD, which has a signal output, which is connected to an input channel of a downstream evaluation unit EV.

In an evaluation mode, a "sum signal" is obtained by adding the output signals from at least two receiver coils of the receiver coil row by means of an adding operation. Preferably, in this adding mode, the output signals of all receiver coils 640A - 640D the receiver coil row synchronously added. The adding operation corresponds in electrical terms to a series connection of the respective receiver coils. For this purpose, it is sufficient if the evaluation unit has a single input channel.

In another evaluation mode ("single signal"), which is not explained in detail here, the output signals of each of the receiver coils are processed separately and assigned to the position of the respective receiver coil in the row. This makes it possible to locate defects across the width of the sensor array.

Claims (9)

A test method for non-destructive material testing of a device under test using a probe array comprising a plurality of probes, wherein the probe array is positioned in proximity to a surface of the device under test; Probe signals from a plurality of probe probes of the probe array are added in an addition operation to a sum signal; and evaluating the summation signal resulting from the addition operation to characterize defects in the device under test in an evaluation operation, characterized in that before the addition operation, the probe signals are subjected to amplitude filtering with respect to a predetermined threshold, whereby probe signals having a signal amplitude below the threshold are filtered out and only probe signals having a signal amplitude greater than or equal to the threshold value of the addition operation are supplied. Test method according to claim 1, characterized in that the test piece and the probe array are moved relative to each other along a direction of movement; A test method according to claim 1 or 2, characterized by a calibration operation, wherein the threshold value is set in dependence on the type of the test object. Test method according to claim 3, characterized in that a noise signal amplitude of a noise signal is determined during the calibration operation on a defect-free reference test specimen or a defect-free specimen section and that the threshold value is set at a predefined distance above the noise signal amplitude, wherein the predefined distance is preferably between 10%. and 20% of the noise signal amplitude. Test method according to one of the preceding claims, characterized in that a probe array with a plurality of eddy current Test probes for performing an eddy current test is used. Test device for non-destructive material testing of a test object with: a probe array having a multiplicity of test probes (P1, P2,... Pn), and an evaluation unit connected to the probe array for evaluating probe signals of the test probes, wherein the evaluation unit has a signal adding device (ADD) and an evaluation device (EV), the signal adding means is configured to perform an addition operation in which probe signals from a plurality of probes of the probe array are added to generate a sum signal (SUM), and the evaluation device is configured to evaluate a summation signal resulting from the addition operation for characterizing defects in the test object in an evaluation operation, marked by an amplitude filter means (AF, AF-AN, AF-DIG) connected between the probes (P1, P2, ... Pn) and the signal adding means (ADD) and configured to relate the probe signals to amplitude filtering prior to the addition operation to subject to a predetermined threshold value, wherein probe signals with a signal amplitude below the threshold filtered out and only probe signals with a signal amplitude greater than or equal to the threshold value of the signal adding device (ADD) are supplied. Test device according to claim 6, characterized in that the test probes (P1, P2, ... Pn) are designed as eddy current probes, as stray flux probes or as GMR probes. Test device according to claim 6 or 7, characterized in that the probe array comprises a plurality of eddy current probes, which has a pathogen arrangement ( 610 ) with at least one electrical conductor for connection to an AC voltage source ( 620 ) as well as a receiver arrangement with at least one receiver coil separate from the exciter arrangement (US Pat. 640 ) having one or more windings, wherein the excitation arrangement has an exciter section ( 615 ), in which one or more conductor sections ( 610-1A . 610-2A ) such that a primary alternating magnetic field (PF) with magnetic field lines (FL) running around the excitation section is generated by the conductor section or the conductor sections of the exciter section when connected to an AC voltage source; and the receiver coil ( 640 ) with respect to the exciter section ( 615 ) is arranged such that the receiver coil of the primary alternating magnetic field of the exciter section is interspersed symmetrically such that a temporal change dφ / dt of the magnetic flux φ of the primary alternating magnetic field by the receiver coil substantially disappears. Test device according to one of claims 6 to 8, characterized in that several or all of the eddy current probes of the probe array have a common excitation arrangement.

Images