FR3003955A1 - DEVICE FOR CONTROLLING A METAL PART AND ASSOCIATED METHOD - Google Patents

Abstract

This device comprises a probe (16) comprising a plurality of measuring coils and a compensation coil. It comprises a selective excitation unit (40) adapted to selectively apply an electrical excitation signal to a first set of measuring coils, and simultaneously to a compensation coil. It comprises a measurement unit (42) capable of selectively receiving an electrical measurement signal from each measuring coil and an electrical compensation signal from each compensation coil, and a calculation unit (46) suitable for determining information representative of the complex impedance of the metal part (12), based on the measurement signal received from said measuring coil and the compensation signal received from said compensation coil.

Description

The present invention relates to a device for controlling a metal part. Such a device is intended for example to measure a local physical quantity representative of the part, such as the conductivity, in order to determine any anomalies in the part. The measured local physical quantity is representative in particular of the state of health of the material, in particular in the thickness of the part. On the basis of this magnitude, it is possible to determine the metallurgical characteristics of the material, and in particular the mechanical strength.

In particular, the device is intended to establish a state map of the part, in the context of the production or maintenance of metal parts, including structural parts of aircraft. This cartography makes it possible to locate, to quantify, and to dimension the anomalies possibly highlighted. The device is particularly applicable to large parts, such as wing panels, aircraft central planes, or aircraft nozzles. It applies in particular to metal parts which have undergone heat treatment, mechanical machining, and / or plating during their production. It also applies for maintenance, for parts that have undergone operational flight conditions. To detect metallurgical anomalies in parts, especially during the production of parts, it is known to perform a visual inspection after BF4 / 5 chromatic anodic oxidation, following a standardized protocol. Visual inspection is done by determining a hue that may depend on the nature of the anomaly.

The visual test, however, can only be performed on parts made from a number of aluminum alloys. For some of these alloys, it is difficult or impossible to apply this test method. In addition, the chromic anodic oxidation is sometimes substituted by anodic oxidation sulfuric, for which the test can no longer be implemented.

To check the state of the part, it is also known to use an eddy current probe comprising a measuring coil, mounted on a manual device. This method applies in particular to determine a possible degradation of the mechanical characteristics following local accidental overheating, or to determine a non-conformal plating thickness.

In this case, a state map of the part can be established manually by performing a grid. The probe is manually moved to each measuring point, where a generally non-continuous measurement is made. Such a method is slow and tedious to implement. In addition, the accuracy of the mapping depends on the skill of the operator. This can in some cases impact the ability to detect anomalies and prevent the determination of the sizing of any anomalies. More generally, the methods used for the control of metal parts are very dependent on the human factor and have a low traceability.

An object of the invention is therefore to provide a device for controlling a metal part, which makes it easy and very reproducible to obtain a precise and detailed mapping of the properties of the part. To this end, the subject of the invention is a device of the aforementioned type, characterized in that it comprises: a probe capable of being applied to the metal part, the probe comprising a plurality of measuring coils spaced apart from each other; others and at least one compensation coil; a selective excitation unit capable of selectively applying an electrical excitation signal to a first set of measuring coils, without applying an electrical excitation signal to the other measuring coils of the plurality of measurement coils, and simultaneously applying the electrical excitation signal to at least one compensation coil; - a measuring unit, adapted to selectively receive an electrical measurement signal from each measuring coil excited by the electrical excitation signal and an electrical compensation signal from each compensation coil excited by the electrical excitation signal; a calculation unit capable of determining information representative of the complex impedance of the metal part, taken opposite each measuring coil, on the basis of the measurement signal received from said measuring coil and the compensation signal received; of said compensation coil. The device according to the invention can comprise one or more of the following characteristics, taken separately or in any technically possible combination: it comprises a control unit of the excitation unit and the measurement unit; , able to excite, in a first time interval of a measurement cycle, the first set of measuring coils and at least one compensation coil, without exciting the other measuring coils of the plurality of measuring coils, and clean to excite, in a second time interval of the measurement cycle following the first time interval of the measurement cycle, a second set of measuring coils distinct from the first set of measuring coils and at least one compensation coil, without exciting the other measuring coils of the plurality of measuring coils, the first set of measuring coils and the second set of measuring coils preferably each comprising at least two coils measuring; the control unit is able to repeat several measurement sequences, each measurement sequence comprising several measurement cycles, each measurement cycle being advantageously defined with time intervals and combinations of measuring coils and different compensation coils; ; the probe comprises at least one measuring coil line spaced from one another, advantageously at least one measuring coil network spaced apart from one another, and arranged in a matrix; the excitation unit is able to generate, for each measuring coil, a multi-frequency excitation electric signal, advantageously comprising at least two distinct frequencies, advantageously at least four distinct frequencies; the calculation unit comprises at least one set for determining a difference between each measurement signal received for a given measuring coil, and the corresponding compensation signal received for a compensation coil excited simultaneously with the given measuring coil. the representative information being obtained from said difference; the calculation unit comprises a conversion set of the information representative of the complex impedance measured from each measuring coil into a value of a physical quantity, advantageously the conductivity of the metal part facing the coil; measuring; the probe comprises a sensor for determining the relative position of the probe on the metal part, the calculation unit comprising a mapping set of a value of a physical quantity obtained from the information representative of the complex impedance measured with respect to each measuring coil, the physical quantity being advantageously chosen as the conductivity of the metal part; the calculation unit comprises a selection set of information representative of the complex impedance obtained with the aid of the measuring coils, the selection set being suitable for selecting the information representative of the complex impedance located in at least one a predetermined domain of a complex impedance plane, advantageously of an optionally truncated angular sector-shaped contour or of a polygonal contour, and to exclude representative information located outside the or of each predetermined domain; - The probe comprises a support wall carrying the measuring coils, the support wall being deformable to the touch to adapt to the local conformation of the metal part. The invention also relates to a method for controlling a metal part, characterized in that it comprises the following steps: - supply of a probe as described above; - Provision of the probe on the metal part, the measuring coils being placed opposite the metal part; - application by the excitation unit of an electrical excitation signal to a first set of measuring coils, without applying an electrical excitation signal to the other measuring coils of the plurality of measuring coils, and simultaneous application from the electrical excitation signal to at least one compensation coil; - receiving by the measuring unit an electrical measurement signal from each measuring coil excited by the electrical excitation signal and an electrical compensation signal from each compensation coil excited by the electrical signal of excitement; determination by the computing unit of information representative of the local complex impedance of the metal part, taken opposite each measuring coil, on the basis of the measurement signal received from said measuring coil and the compensation signal received from a compensation coil excited simultaneously with said measuring coil.

The method according to the invention may comprise one or more of the following characteristics, taken separately or in any technically possible combination: it comprises a first excitation step, in a first time interval of a cycle of measurements, of the first set of measuring coils and at least one compensation coil, without exciting the other measuring coils of the plurality of measurement coils, and at least a second excitation step, in a second time interval of the measurement cycle consecutive to the first time interval, of a second set of measuring coils distinct from the first set of measuring coils and a compensation coil, without exciting the other measuring coils of the plurality of measuring coils ; it comprises, for each measuring coil, the application by the excitation assembly of an electrical multi-frequency excitation signal, advantageously comprising at least two distinct frequencies, advantageously at least four distinct frequencies; the calculation step comprises converting the information representative of the complex impedance measured from each measurement coil into a value of a physical quantity, advantageously the conductivity of the metal part facing the measuring coil. ; the calculation step comprises selecting, among the information representative of the complex impedance obtained using the measuring coils, representative information situated in a predetermined domain of a complex impedance plane, the or each domain advantageously being in the shape of an optionally truncated angular sector or polygonal contour, and the exclusion of representative information located outside the predetermined domain.

The invention will be better understood on reading the description which follows, given solely by way of example, and with reference to the appended drawings, in which: FIG. 1 is a diagrammatic front view of a first control device according to the invention, during the control of a metal part; FIG. 2 is an enlarged view of the measuring probe of the device represented in FIG. 1; FIG. 3 is a view from below of the arrangement of the measuring coils and compensation coils in the probe of FIG. 2; FIG. 4 is a table illustrating the relative position of each measuring coil on the probe; FIG. 5 is a schematic representation of the electrical connection circuit between the measuring and compensation coils, an excitation unit for the coils, a measurement unit, a control unit and a calculation unit, in the device of FIG. figure 1 ; FIG. 6 is a table illustrating the multiplexing of the measurements carried out with the aid of the device of FIG. 1; FIG. 7 is a view similar to FIG. 3, illustrating a first measuring step during a first time interval; FIG. 8 is a view similar to FIG. 7, illustrating a second measuring step during a second time interval; FIG. 9 is a schematic view of the complex impedance plane illustrating the measurements made by the device of FIG. 1, and among the measurements made, those which are selected to calculate the value of a physical quantity associated with the metal part. ; FIG. 10 is a view illustrating a calibration function between the value of information representative of the local impedance of the metal part measured by a measurement coil and the value of the corresponding physical quantity; FIG. 11 is a two-dimensional view of a map of a metal part made using the device according to the invention; - Figure 12 is a view similar to Figure 3 of the probe of a second device according to the invention. A first device 10 according to the invention is illustrated schematically in FIG. 1. Such a device 10 is intended to locally measure a physical quantity representative of a metal part 12 visible in FIG. 1, in a plurality of points on the part 12 , to advantageously obtain a map of the value of the physical quantity on the metal part 12. The physical quantity is for example the electrical conductivity of the part. In a variant, the conductivity is measured on at least two layers of metallic materials, for example in the case of a plating. By "plating" is meant the deposition of an almost pure aluminum foil on the surface of an aluminum alloy sheet. The control of the material health is then carried out through each of the two layers. The electrical conductivity that can be measured by the device 10 according to the invention is, for example, between 1 MS / m and 35 MS / m, advantageously in a range of values of 20 MS / m lying in this range, and whose lower limit is defined by the compensation block associated with the measurement, with an accuracy of at least 0.2 MS / m. The part 12 has a surface 14 on which the measurement is made. The part 12 advantageously has large dimensions, in particular depending on the size of the probe. The mechanical part 12 is in particular an element of a flying platform, such as a civil or military aircraft. This piece is for example a wing panel, a central plane of aircraft, or an aircraft nose.

In a first application, the piece 12 has for example undergone a heat treatment, a mechanical machining, and / or a plating, during its manufacture, and is tested during the manufacture or at the end of it, to determine by For example, a local overheating of the part 12. In another application, the part 12 has been used under operational conditions in flight and is tested after a predetermined number of hours of flight to determine for example a possible degradation of the mechanical characteristics following local accidental overheating or improper plating thickness. The metal part 12 for example formed from an aluminum alloy. With reference to FIG. 1, the device 10 comprises a probe 16 intended to be applied on the outer surface 14 of the metal part 12, and an assembly 18 for driving the probe 16 and for processing the data coming from the probe 16 connected to the probe 16 by a wired or wireless connection link 19 via a wireless wiring or transmission block 23. As shown in FIGS. 2 to 3, the probe 16 comprises a housing 20, a plurality of coils of measurement 22 forming a network 24 visible in Figure 3, and a wall 26 for supporting measuring coils 22. The probe 16 further comprises at least one compensation coil 28, and for each compensation coil 28, a block 30 compensation metal (or wedge) on which the compensation coil 28 is supported.

The housing 20 is adapted to be gripped by the hand of an operator or by a robot to be placed on a region of the part 12. It delimits an internal housing 29 receiving in this example each compensation block 30 and each compensation coil 28. The inner housing 29 is closed outwardly by the support wall 26.

In this example, the support wall 26 is deformable to the touch to adapt to the local conformation of the part 12. It is for example formed by a block of elastic material, such as silicone. The support wall 26 can thus be deformed to locally adopt a radius of curvature of less than 100 mm, in particular of the order of 50 mm.

The support wall 26 has an outer face 31 intended to come into contact with the part 12 to be characterized. The area of the outer face 31 is for example greater than 8 mm 2, and is for example between 2 mm 2 and 2000 mm 2. Each measuring coil 22 is carried by the support wall 26. In this example, each measuring coil 22 has a substantially cylindrical shape, of axis substantially perpendicular locally to the outer face 31.

Each coil 22 comprises a cylindrical casing 32 having an outer surface 34 of contact with the part 12 to be characterized. In this example, each coil 22 is embedded at least partially in the support wall 26. The outer surface 34 of each coil 22 is flush with the outer face 31 of the support wall 26. Thus, the air gap between each coil 22 and the surface 14 of the part 12 is minimized during the application of the probe 16 against the surface 14 of the part 12. The number of measuring coils 22 carried by the probe 16 is greater than 2, advantageously greater than 16 and is particularly 16 to 256. Advantageously, the measuring coils 22 all have identical dimensions. Each measuring coil 22 has for example a height of between 1 mm and 20 mm and a diameter of between 1 mm and 20 mm. The measuring coils 22 are spaced apart from each other. The minimum distance separating each pair of adjacent coils 22 is greater than 0.2 times the minimum diameter of each coil 22, and is preferably between 0.4 times and 0.6 times the minimum diameter of each coil 22. the example shown in Figure 3, the network 24 of measurement coils 22 is regularly organized in the form of a matrix organized in rows and columns. It comprises two parallel lines 36 each comprising a plurality of measurement coils 22, here nine measurement coils 22 per line. The measuring coils 22 of two adjacent lines 36 are staggered, to obtain a regular coverage of the measured part 12. Alternatively, the network 24 may comprise more than two lines 36, for example four lines 36 or even sixteen lines 36 to form for example a 16x16 matrix consisting of 256 elements. In another variant, the network 24 is not a matrix but a single row of measuring coils 22. The probe 16 comprises at least one compensation coil 28, advantageously at least two compensation coils 28. In the example As shown in FIG. 2, each compensation coil 28 and the associated compensation block 30 are received in the inner housing 29 of the housing 20. Each compensation coil 28 is of the same size and structure as the measuring coils 22. 35 Alternatively, the compensation coils are housed in the wiring block 23.

Each compensation coil 28 is applied to the compensation block 30. Each compensation coil 28 and its compensation block 30 are disposed above each measuring coil 22 and the support wall 26, when the support wall 26 is directed downwards in the position of Figure 2.

In addition, each compensation coil 28 and its associated compensation block 30 are arranged laterally away from the measuring coils 22 at the periphery thereof. In the example of FIG. 3, the compensation coils 28 are situated laterally on either side of the lines 36 of measuring coils 22, positioned transversely between the two lines 36 with respect to the axis of each line. 36. Each compensation block 30 is made of a metal of known conductivity having a conductivity of between 1 MS / m and 35 MS / m. Each compensation block 30 is advantageously interchangeable in a reversible manner to adapt to the desired measurement range. The conductivity value of the compensation block 30 defines the minimum of the measuring range of the probe 16. It is possible to provide two identical compensation blocks 30 or two different compensation blocks 30 so that they can be switched from one range to another. from one measure to another without having to change these blocks 30.

In an advantageous variant, the probe 16 carries at least one sensor 38 for determining its position on the part 12. The sensor 38 is for example an encoder capable of measuring at least two dimensions the displacement of the probe 16 on the part 12. As illustrated by FIGS. 1 and 5, the driving and processing assembly 18 comprises a selective excitation unit 40 for each coil 22, 28, adapted to apply to each coil 22, 28 an oscillating electrical signal of excitation, and a measurement unit 42, adapted to selectively receive an electrical measurement signal from each coil 22, 28. The assembly 18 further comprises a control unit 44 of the excitation unit 40 and the measurement unit 42, for performing a multiplexing between the measurements, and a unit 46 for calculating information representative of the complex impedance measured by each coil 22, to determine a local physical quantity on the part 12. As illustrated by FIG. 5, the excitation unit 40 comprises a source 50 for generating the electrical excitation signal, an input selective switch 52, connected to the source 50, and for each coil 22, 28, an electrical path 54 selective supply of the coil 22, 28 selectively connecting the coil 22, 28 to the selective input switch 52. For simplicity, only some paths 54 are shown in Figure 5. The source 50 is for example a source of electric current oscillating. It is suitable for generating the electrical excitation signal at at least one excitation frequency greater than 10 Hz, and in particular between 50 kHz and 500 kHz. Advantageously, the source 50 is adapted to generate an electrical multi-frequency excitation signal, comprising simultaneously or consecutively several excitation frequencies, advantageously from two to four excitation frequencies.

The excitation frequencies are for example 60 kHz, 120 kHz, 240 kHz, and 480 kHz. The selective input switch 52 is connected to the source 50 and to the control unit 44. It comprises a plurality of input switches 56, arranged in parallel and each interposed between the source 50 and a particular coil 22, 28. Each switch 56 is adapted to be selectively switched by the control unit 44 between a closed power supply configuration of the particular coil 22, 28 by the electrical excitation signal and an open disabling configuration. coil 22, 28.

Each electrical supply path 54 connects an input switch 56 advantageously to a single coil 22, 28. As will be seen below, the excitation unit 40 is clean, in a given time interval T1, T2, selectively applying an electrical excitation signal to a first set of measuring coils 22, comprising at least one measuring coil 22, without applying the electrical excitation signal to the other measuring coils 22. The unit 40 is suitable for simultaneously applying, in the given excitation interval, the electrical excitation signal to at least one compensation coil 28. With reference to FIG. 5, the measurement unit 42 comprises, for each coil 22, 28, a electrical output path 60 specific to this coil 22, 28, at least one selective output switch 62, receiving the electrical output paths 60 and at least one measurement channel 64, connected to the output switch 62. For simplicity, only certain paths 60 have been represented. Each output electrical path 60 selectively connects a coil 22, 28 to an output selective switch 62.

Each output selector switch 62 has a plurality of output switches 63 arranged in parallel and selectively interposed between an electrical output path 60 and a channel 66, 68 for measuring a channel 64. Each output switch 63 is suitable for selectively switched by the control unit 44 between a closed measurement configuration of the signal received from the particular coil 22, 28 when the coil 22, 28 is energized by the electrical excitation signal and an open-circuit configuration. the measurement from the coil 22, 28. Each measurement channel 64 includes a first channel 66 for selective connection to a measurement coil 22 through an output switch 63, for receiving an electrical measurement signal from the coil 22 when the measuring coil 22 is excited by the electrical excitation signal. Each channel 64 further includes a second path 68 for selectively connecting to a compensation coil 28 for receiving an electrical compensation signal from the compensation coil 28 when the coil 28 is energized by the electrical excitation signal. . In the example shown in FIG. 5, the measurement unit 42 comprises two selective output switches 62 and two associated measurement channels 64. A first output switch 62 is connected to each first channel 66 of each channel 64 on the one hand, and to the plurality of measurement coils 22 on the other hand, to electrically supply the first channel 66 with an electrical measurement signal coming from selectively from each measuring coil 22. A second switch 62 is connected to each second channel 68 of each channel 64 on the one hand, and to the plurality of compensation coils 28 on the other hand, to electrically power the second channel 68 with an electrical compensating signal coming selectively from each compensation coil 28. The control unit 44 is able to perform a multiplexing between the measurements made by the coils 22. It contains a module 70 for the selective control over time of the switches. 56 and output 63 of the switches 52, 64, according to an activation table 72 illustrated for example in Figure 6. The module 70 is suitable for realizing man repetitive sequence of measurement sequences comprising at least one measurement cycle. Each measurement cycle comprises a plurality of consecutive time intervals T1, T2 each forming a measurement window.

The module 70 is configured to selectively switch each of the switches 56, 63 according to the excitation scheme and the measurement scheme defined in the activation table 72. The activation table 72 defines, for each time interval Ti, T2 of a measurement cycle, a set of measuring coils 22 intended to be selectively powered by the electrical excitation signal, without power supply of the other measurement coils 22 by this signal (columns E1 and E2 in FIG. 6) . The activation table 72 further defines, for each time interval T1, T2, the compensation coil or coils 28 intended to be supplied simultaneously by the electrical excitation signal (column E3 in FIG. 6). The activation table 72 further defines on which channel C1 or C2 are oriented the signal from each measuring coil 22 and the signal from each compensation coil 28 excited simultaneously during the measurement interval T1, T2. Advantageously, each set of measuring coils 22 excited at a given time interval T1, T2 comprises between one and four measuring coils 22, spaced from each other, and separated from each other by at least one measuring coil 22. excited, to limit interference. Simultaneously, the number of compensation coils 28 excited is at least equal to one or a compensation coil for all measuring coils and is at most equal to the number of measuring coils or a compensation coil for a measuring coil. The number of compensation coils 28 excited is advantageously equal to a quarter of the number of measuring coils 22, each compensation coil 28 being associated with at most four measuring coils 22. In the example of FIG. 6, each set of coils The measuring circuit 22 energized a given time interval T1, T2 comprises two measuring coils 22, a compensation coil 28 being excited simultaneously. Each time interval Ti, T2 given is proportional to the excitation frequency. Advantageously, it is worth five periods of the signal at the lowest frequency. The number of time intervals T1, T2 necessary for the excitation of all the measuring coils 22 of the probe 16 in a measuring cycle is between 2 and 32. Thus, the control unit 44 is able to drive the excitation unit 40 for exciting, in a first time interval T1, a first set of measuring coils 22 and at least one compensation coil 28, without exciting the other measuring coils 22 of the probe 16.

The control unit 44 is able to drive the excitation unit 40 to excite, in a second time interval T2 consecutive to the first time interval Ti, a second set of measurement coils 22 distinct from the first set of coils of measurements 22 and advantageously, a compensation coil 28 distinct from that excited in the first time interval T1, without exciting the other measuring coils 22 or the other compensation coils 28 of the probe 16.

The control unit 44 is also able to control each switch 63 of each output switch 62, to direct the measurement signal collected on each measuring coil 22 excited during a time interval T1, T2 to the first channel 66 of a channel 64, and to direct the compensation signal collected on a corresponding compensation coil 28 to the second channel 68 of the same channel 64.

The control unit 44 is thus able to repeat several measurement sequences, themselves comprising several measuring cycles, each measuring cycle being able to be defined with time intervals T1, T2 and combinations of measuring coils 22 and coils. 28 different compensation. This ability to sequence measurement cycles with different multiplexings allows you to analyze a part simultaneously with several resolutions by changing the measuring range at each measurement sequence. Indeed, on the one hand, the multiplexing makes it possible to fix the association of the measuring coils with the compensation coils and their reference shim and on the other hand the conductivity value of the reference shim defines the low value of measuring range. So each sequence is associated with a new reference block adapted for a different conductivity value. The calculation unit 46 comprises, for each channel 64, a set 80 for determining the difference between the measurement signal received from a given measurement coil 22 on the first channel 66, and the compensation signal received at from a compensation coil 28 associated with the second channel 68, to obtain information representative of a complex impedance of the metal part 12, taken at the level of the measurement coil 22 given. The calculation unit 46 furthermore comprises a set 81 for selecting the representative information obtained, a set 82 for converting each selected representative information into an associated physical quantity, such as the conductivity, and advantageously a set 84 a mapping of the value of the physical quantity on the metal part 12 to be checked. The determination unit 80 is able to subtract, in a complex plane, the compensation signal from the measurement signal.

The representative information 86 obtained is thus in the form of a determined complex value, associated with each measurement coil 22, in a complex impedance plane illustrated schematically in FIG. 9. The selection set 81 is suitable for selecting among the representative information 86 obtained for each measuring coil 22, those which are located in at least one predetermined domain 88, 90, 92 of the complex impedance plane, and to eliminate the representative information 86 that is located outside of each domain 88, 90, 92. The domains 88, 90, 92 are defined, for example, as a function of a given phase, and / or as a function of a given module, and / or as a function of a real part or of a given part. imaginary part given. For example, a first domain 88 has an angular sector shape delimited by two boundary phases P1 and P2. A second domain 90 has a truncated angular sector shape (or tart portion), delimited by two boundary phases P3 and P4 and two boundary modules M1 and M2 in the complex impedance plane. A third domain 92 is of polygonal shape and is delimited by two boundary real limit values X1 and X2 and by two limiting imaginary part values Y1 and Y2. The limits of the domains 88, 90, 92 are advantageously dependent on the type of physical quantity calculated from the measured representative information 86, for example the conductivity and / or the range of values of the physical quantity considered, and / or the position of the probe 16 on the part 12. In this example, only the representative information 86 contained in one or more domains 88, 90, 92 (represented in solid lines in FIG. 9) are retained by the selection assembly 81, the representative information 86 located outside each domain 88, 90, 92 (shown in dashed lines in FIG. 9) being eliminated by the selection set 81. The conversion assembly 82 includes a unit capable of calculating a given physical quantity value. associated with each measuring coil 22, based on each representative information 86 from the selection set 81, and a predetermined calibration function 94, illustrated by the curve of FIG. 10. The predetermined calibration function 94 is established for example by measuring the representative information 86 obtained by means of the probe 16 on at least one calibration piece having a known physical magnitude value. Advantageously, this function 94 is determined from a plurality of measurements 96 performed on several calibration pieces, for example between two and eight different calibration pieces. Linear or polynomial extrapolation is used to determine the calibration function 94 connecting each measured representative information 86 to the value of the associated physical quantity. In the case of conductivity, the calibration function 94 is established between the imaginary part Y of the representative information 86 measured, balanced in phase so as to adjust the complex measurement signal so as to obtain a deflection along the vertical axis proportional to the change in conductivity, and the conductivity, expressed in MS / m. The mapping set 84 is capable of associating each representative information 86 measured with the aid of a measuring coil 22, and the value of the corresponding physical quantity, at the precise position of the measurement coil. 22 on the metal part 12 during the measurement. For this purpose, the set of establishment 84 is able to connect the representative information 86 and the value of the corresponding physical quantity, to a position of a reference associated with the probe 16 on the metal part 12, as determined by the sensor 38, and at a relative position of the measuring coil 22 considered on the probe 16, determined in the coordinate system associated with the probe 16 by means of a table 98 for positioning the probe 16 in the matrix of the network 24 shown schematically in FIG. 4. The set of establishments 84 is able to create a graphic representation illustrating the value of the physical quantity as a function of the position on the metal part 12 to be checked, for example by associating particular colors. at a plurality of ranges of values of the physical quantity. The graphic representation thus obtained has pixels which each correspond to a value of the physical quantity determined by a particular measuring coil 22 of the probe 16, in a given position of the probe 16 on the part 12 to be checked. Such mapping is for example illustrated in Figure 11, and allows to establish the presence, size, position, and the number of possible defects 100 on the metal part 12 to control.

The resolution of the probe 16 (size of a pixel) is proportional to the number of measuring coils 22 and inversely proportional to the size of these coils 22. A method of controlling a metal part 12 according to the invention will now be described. Initially, the device 10 according to the invention is placed in the vicinity of the metal part 12. The probe 16 is then grasped by an operator or by a robot and is applied against a surface 14 of the metal part 12 on a given region of this 12. Advantageously, the support wall 26 deforms to match the local conformation of the metal part 12.

Each measuring coil 22 then occupies a given position on the metal part 12, determined by the sensor 38, and by the positioning table 98 of the coil 22 on the probe 16. The outer surface 34 of each measuring coil 22 is placed in contact with or near the metal part 12.

Then, the control and processing unit 18 is activated to perform a measurement cycle. The source 50 of the excitation unit 40 is then turned on. In a first measurement interval Ti, the control unit 44 drives the excitation unit 40 to excite a first set of measurement coils 22, and at least a first compensation coil 28, with an electrical excitation signal. generated by the source 50, without exciting the other measuring coils 22 of the probe 16, nor the other compensation coils 28 of the probe 16. For this purpose, the control unit 44 controls the switches 56 of the selective switch. input 52 to selectively connect each measuring coil 22 of the first set to the source 50 and to electrically isolate the other measuring coils 22 from the source 50. Simultaneously, the control unit 44 drives the switches 56 of the selective input switch 52 to electrically connect at least one compensation coil 28 associated with the first set of measurement coils 22 to the source 50, and to electrically isolate the other compensation coils 28.

In the example illustrated in FIG. 7, the first set comprises two measuring coils 22 situated respectively on a first line 36 and on a second line 36 at opposite axial positions on these lines 36, in order to limit electromagnetic interference between the coils. The other measuring coils 22 are not electrically powered.

A compensation coil 28 is further powered by the electrical excitation signal, while the other coil 28 is not electrically powered. The electrical excitation signal generates by induction, on the metal part 12 opposite each excited measuring coil, eddy currents whose intensity depends in particular on the local impedance of the metal part 12 opposite the coil measuring 22.

The eddy currents generate in turn by induction an electrical measurement signal depending in particular on the local impedance of the metal part 12. This electrical measurement signal is picked up by each measuring coil 22. Likewise, the electrical signal of excitation generates by induction, on the compensation block 30, eddy currents whose intensity depends in particular on the local impedance of the block 30 facing the compensation coil 28. The eddy currents generate in turn by induction a electrical compensation signal depending in particular on the local impedance of the block 30. This compensation electric signal is picked up by the compensation coil 28.

The control unit 44 then controls each switch 63 of each output switch 62 to electrically connect a first channel 66 of an output channel 64 to each measurement coil 22 excited via an output path 60, through a closed output switch 63. It further controls each switch 63 of each output switch 62 to electrically connect a second channel 68 of the output channel 64 to the compensation coil 28 excited simultaneously with the measuring coil 22. In the example of FIG. first coil 22 of the first set of measuring coils 22 is connected to the first channel 66 of a first channel 64, and the second coil 22 of the first set of measuring coils is connected to the first channel 66 of a second channel 64 for collecting an electrical measurement signal on this first channel 66. The associated compensation coil 28 is connected to each second channel 68 of each channel 64 to collect an electrical compensation signal on this second channel 68.

Then, the determination unit 80 makes the complex difference between the electrical measurement signal and the electrical compensation signal, in order to obtain an information item 86 representative of the local impedance of the piece 12 facing each excited measurement coil 22. Then, in a second measuring interval T2, the previous operations are repeated identically for a second set of measuring coils 22 distinct from the first set of measuring coils 22 excited during the first measuring interval T1, as illustrated by FIG. In this example, two measuring coils 22 separate from the coils 22 excited during the first measuring interval T1 are excited during the second measuring interval T2.

Similarly, a compensation coil 28 distinct from that excited during the first measurement interval T1 is excited during the second measurement interval T2. These operations are repeated identically for a measuring cycle comprising a plurality of successive measurement intervals Tn, by exciting at each Tn interval at least one measurement coil 22 distinct from those excited at the previous intervals. Once the measurement cycle has been performed, when all the measurement coils 22 have been excited, the selection set 81 retains the representative information 86 located in the predetermined selection areas 88, 90, 92, and excludes the representative information 86 located outside these areas. Then, the conversion unit 82 converts each representative information 86 by the selection set 81 into a value of a given physical quantity, based on the predetermined calibration function 94. The mapping set 84 can then associate each value of a given physical quantity, determined from the measurement signal collected by a given measurement coil 22, with the position, on the metal part 12, of this same measuring coil 22 during the excitation of the measuring coil 22. A mapping of the region of the metal part 12 covered by the probe 16 can then be generated in a table of a computer spreadsheet and / or displayed on a screen. Advantageously, the probe 16 is then moved to be placed on an adjacent region of the part 12 and the previous operations are reproduced for a new measuring cycle. In a variant, it is the part 12 to be controlled which is displaced with respect to the probe 16 or else the control is carried out only on a single zone of the part 12 without displacement of the probe 16. A complete cartography, such as illustrated by Figure 11 can then be obtained, when the entire area of interest has been scanned. The device 10 according to the invention thus makes it possible to obtain a fast, precise and reproducible control of large parts 12, without having to perform preliminary preparation of the part 12. It provides a direct visualization of each region of the part 12 scanned by the probe 16, and allows the detection, quantification, and sizing of possible defects 100 in this room.

The support wall 26 being flexible, it is possible to measure on left metal parts 12, for example having a radius of local curvature of the order of 50 mm. The multiplexing performed using the control unit 44 makes it possible to scan and compensate in a single step a large number of adjacent measurement coils 22, for example more than ten adjacent measuring coils, and to excite each of the coils. measurement coils 22 using a multi-frequency signal, which can include in particular up to four simultaneous frequencies. The device 10 directly calculates the value of the measured physical quantity, expressed in a suitable and standardized measurement unit, for example in MS / m for the conductivity, and in micrometers for the plating thickness. This result is obtained thanks to the presence of a conversion unit 82 capable of directly converting the signal obtained for example in the form of a measured voltage in the right unit.

The phased, multi-frequency analysis of the measurement signals allows selection of the representative information collected to sort out those that are relevant for the analysis of possible anomalies and eliminate measurement artifacts. The device 10 is also able to perform an accurate mapping of the value of the physical quantity on the metal part 12, which can be easily restored in the form of a table in a spreadsheet, and / or in the form of a display graphic. The matrix probe 16 of a second device 210 according to the invention is illustrated in FIG. 12. Unlike the probe 16 of the first device 10, this probe 16 comprises at least a third line 36 of measuring coils 22, advantageously disposed parallel to a first line 36 of measuring coils 22 and parallel to a second line 36 of measuring coils 22. The probe 16 shown in Figure 12 further comprises at least two pairs of compensation coils 28 located laterally on the side and other lines 36 of measuring coils 22.

The operation of the probe 16 shown in FIG. 12 is, moreover, similar to that of the probe 16 shown in FIG. 1. However, the probe 16 of FIG. 12 is able to collect a number of measurements that are higher than that of the probe 16 shown in FIG. 1 during the same measurement cycle.

Claims (15)

CLAIMS1.- Device (10; 210) for controlling a metal part (12), characterized in that it comprises: - a probe (16) adapted to be applied to the metal part (12), the probe ( 16) having a plurality of measuring coils (22) spaced from one another and at least one compensation coil (28); a selective excitation unit (40) capable of selectively applying an electrical excitation signal to a first set of measuring coils (22) without applying an electrical excitation signal to the other measurement coils (22); the plurality of measuring coils (22), and adapted to simultaneously apply the electrical excitation signal to at least one compensation coil (28); a measurement unit (42) adapted to selectively receive an electrical measurement signal from each measuring coil (22) excited by the electrical excitation signal and an electrical compensation signal from each compensation coil (28). excited by the electrical excitation signal; a calculation unit (46) capable of determining representative information (86) of the complex impedance of the metal part (12), taken opposite each measuring coil (22), on the basis of the measurement signal received from said measuring coil (22) and the compensation signal received from said compensation coil (28). 2.- Device (10; 210) according to claim 1, characterized in that it comprises a unit (44) for controlling the excitation unit (40) and the measuring unit (42), clean to excite, in a first time interval (Ti) of a measurement cycle, the first set of measuring coils (22) and at least one compensation coil (28), without exciting the other measuring coils (22) of the plurality of measuring coils (22), and adapted to excite, in a second time interval (T2) of the measurement cycle consecutive to the first time interval (Ti) of the measuring cycle, a second set of measuring coils (22) separate from the first set of measuring coils (22) and at least one compensation coil (28), without exciting the other measuring coils (22) of the plurality of measuring coils (22), the first set of measuring coils (22) and the second set of measuring coils (22) preferably each comprising at least two measuring coils (22). 3.- Device (10; 210) according to claim 2, characterized in that the control unit (44) is able to repeat several measurement sequences, each measurement sequence comprising several measurement cycles, each measurement cycle being advantageously defined with time intervals (Ti, T2) and combinations of measuring coils (22) and different compensation coils (28). 4.- Device (10; 210) according to any one of the preceding claims, characterized in that the probe (16) comprises at least one line (36) of measuring coils (22) spaced from each other, preferably at at least one network (24) of measurement coils (22) spaced from each other and arranged in a matrix. 5.- Device (10; 210) according to any one of the preceding claims, characterized in that the excitation unit (40) is able to generate, for each measuring coil (22), an electrical signal of multi-frequency excitation, advantageously comprising at least two distinct frequencies, advantageously at least four distinct frequencies. 6. Device (10; 210) according to any one of the preceding claims, characterized in that the computing unit (46) comprises at least one set (80) for determining a difference between each measurement signal received. for a given measuring coil (22), and the corresponding compensation signal received for a compensation coil (28) excited simultaneously with the given measuring coil (22), the representative information (86) being obtained from said difference. 7.- Device (10; 210) according to any one of the preceding claims, characterized in that the computing unit (46) comprises a set (82) for converting the representative information (86) of the impedance complex measured from each measuring coil (22) to a value of a physical quantity, advantageously the conductivity of the metal part facing the measuring coil (22). 8.- Device (10; 210) according to any one of the preceding claims, characterized in that the probe (16) comprises a sensor (38) for determining the relative position of the probe (16) on the metal part ( 12), the computing unit (46) having a mapping set (84) of a value of a physical quantity obtained from the representative information (86) of the measured complex impedance facing each measuring coil (22), the physical quantity being advantageously chosen for the conductivity of the metal part (12). 9.- Device (10; 210) according to any one of the preceding claims, characterized in that the computing unit (46) comprises a set (81) for selecting the representative information (86) of the complex impedance obtained. using the measuring coils (22), the selection set (81) being adapted to select the representative information (86) of the complex impedance located in at least one predetermined domain (88, 90, 92) d a complex impedance plane, advantageously of an optionally truncated angular sector-shaped contour or a polygonal contour, and excluding the representative information (86) located outside the or each predetermined domain (88, 90, 92). 10.- Device (10; 210) according to any one of the preceding claims, characterized in that the probe comprises a wall (26) of support carrying the measuring coils (22), the support wall (26) being deformable to the touch to adapt to the local conformation of the metal part (12). 11. A method of controlling a metal part (12), characterized in that it comprises the following steps: - providing a probe (16) according to any one of the preceding claims; - Provision of the probe (16) on the metal part (12), the measuring coils (22) being placed facing the metal part (12); - application by the excitation unit (40) of an electrical excitation signal to a first set of measuring coils (22), without applying an electrical excitation signal to the other measuring coils (22) of the a plurality of measuring coils (22), and simultaneously applying the electrical excitation signal to at least one compensation coil (28); - receiving by the measuring unit (42) an electrical measurement signal from each measuring coil (22) excited by the electrical excitation signal and an electrical compensation signal from each compensation coil ( 28) excited by the electrical excitation signal; determination by the computing unit (46) of representative information (86) of the local complex impedance of the metal part (12), taken opposite each measurement coil (22), on the basis of the measurement signal received from said measuring coil (22) and the compensation signal received from a compensation coil (28) excited simultaneously with said measuring coil. 12. A process according to claim 11, characterized in that it comprises a first step of excitation, in a first time interval (T1) of a measurement cycle, of the first set of measuring coils (22) and at least one compensation coil (28), without exciting the other measuring coils (22) of the plurality of measurement coils (22), and at least one second excitation step, in a second time interval ( T2) of the measurement cycle consecutive to the first time interval (T1), of a second set of measuring coils (22) separate from the first set of measuring coils (22) and a compensation coil (28), without exciting the other measuring coils (22) of the plurality of measuring coils (22). 13.- Method according to any one of claims 11 or 12, characterized in that it comprises, for each measuring coil (22), the application by the excitation assembly (40) of an electrical signal multi-frequency excitation, advantageously comprising at least two distinct frequencies, advantageously at least four distinct frequencies. 14. A method according to any one of claims 11 to 13, characterized in that the calculation step comprises the conversion of the representative information (86) of the complex impedance measured from each measuring coil (22). ) to a value of a physical quantity, advantageously the conductivity of the metal part (12) facing the measuring coil (22). 15.- Method according to any one of claims 11 to 14, characterized in that the calculation step comprises selecting, from the representative information (86) of the complex impedance obtained using the measuring coils ( 22), representative information (86) situated in a predetermined domain (88, 90, 92) of a complex impedance plane, the or each domain being advantageously of contour in the form of an optionally truncated angular sector or of a polygonal contour, and excluding representative information (86) outside the predetermined domain (88, 90, 92). 25 30