EP1940564B1 - Sensor apparatus for detecting electromagnetically detectable conveyed goods and sorting apparatus having such a sensor apparatus - Google Patents

Abstract

The invention relates to a sensor apparatus (14) having a plurality of sensors (100) for detecting electromagnetically detectable conveyed goods (15, 16), having an associated conveying device (12) which moves conveyed goods (13) past the sensors (100) in a conveying plane and in a direction of movement (116), having a device (102) for generating an alternating electromagnetic field, wherein the sensors (100) each have at least one pair of detector coils (104, 106) which are connected to an evaluation device (20) for determining a differential signal between the coils (104, 106) in a pair for the purpose of detecting conveyed goods (15) which deform the alternating field on the basis of their material properties, and is characterized in that the two detector coils (104, 106) in the pair are arranged in such a manner that they have cross-sectional areas which are parallel to the conveying plane and have centroids (M, M') which have been displaced relative to one another, and the connecting line (L) between the centroids (M, M') is oblique to the direction of movement (116) of the conveyed goods (13). Furthermore, the invention relates to an apparatus (10) for the automated sorting of electromagnetically detectable fractions of a stream of conveyed goods (13) using such a sensor apparatus (14).

Description

The invention relates to a sensor device according to claim 1 and a sorting device with such a sensor device according to claim 17.

In many areas of recycling, an automated sorting of the recycling material as a conveyed material is required. It should not only process the largest possible Fördergutmengen per time, but the sorting should also be done with high yield and low error rate. It may be in the conveyed example, waste glass, in which there are metal fractions, such as bottle caps or other bottle or glass closures. It may be in the conveyed example also shredded in a shredded scrap cars with fractions of various metals or other valuable materials that are to be made available for recycling. Also a sorting of garbage would be a possible application, eg to sort out an aluminum fraction. Furthermore, it may also be conveyed with different mineralogical fractions, which have different electromagnetic properties and should be sorted for further processing. Not conclusively as applications are also mentioned the sorting of metal residues in wood recycling fractions in the fiberboard industry and the finding of metals in food streams in bulk form.

Various automated devices and methods are known in the art that perform these sorting tasks. For this purpose, the devices have sensors, through whose sensory monitored area the conveyed material is moved. The parts to be sorted out are detected by the sensors designed to be suitable for the sorting criteria, and a separating device is activated by means of the sensor information in order to selectively separate a part which is to be sorted out from the conveyed material. For glass, e.g. is known to make a sorting on the glass color with optical sensors that detect the glass color, and e.g. Separate brown glass from white and green glass.

In particular, devices and methods are known which separate individual fractions according to their distinguishable electromagnetic properties. For this purpose, an electromagnetic alternating field is used, through which the material to be sorted is moved. A change of the alternating field through one of the parts is detected, and then the part sorted out. Such a sorting device with a generic sensor device discloses the EP 0 353 457 B1 , The frequency of the alternating field can be freely selected within wide ranges, eg with a frequency between 5 kHz and 2 MHz.

The sensors of the sensor device shown there consist of two detector coils wound in opposite directions, in which an external alternating field induces equal but opposite alternating voltages. In the undisturbed state, the AC voltages cancel each other out with suitable subtraction exactly to zero. However, if metallic parts reach the measuring range of the detector coils, the homogeneous alternating field is superimposed by an inhomogeneous field induced by the alternating field in the metal parts. This represents a change of the alternating field for the sensors. The alternating field, which is initially generated as homogeneously as possible, now exhibits an inhomogeneous disturbance due to an induced magnetic field. As a consequence, no equal-sized, canceling alternating voltages are now induced in the two detector coils of a sensor, but a difference in value results in a signal value that deviates significantly from zero.

The spatial resolution of these sensors of the generic sensor device is determined by the size of the individual sensors or by the size of the coil pairs contained therein. Also, the separator controlled by the sensors, e.g. a series arrangement of exhaust nozzles, can only be controlled in this limited spatial resolution. This is particularly disadvantageous if the parts to be sorted may be smaller than the sensors, e.g. if the conveyed material is present as granules with a small grain size. It can then be e.g. a non-metal closely adjacent to a metal part accidentally be sorted out by the driven exhaust nozzle with. It creates an undesirable Übersortierung. To achieve a higher resolution, the sensors would have to be downsized. This is expensive to manufacture and would make the sensors more expensive. In addition, the detection range of the sensors would decrease with increasing miniaturization because the effective area for measuring changing electromagnetic field lines would become smaller. Furthermore, the generic sensor device can not correctly detect large metal parts that extend over a plurality of adjacent sensors. Under unfavorable circumstances, the blow-off nozzles are activated incorrectly or not at all.

A sensor device with improved spatial resolution shows the W02005 / 028129 , There are, however, the coils of a coil pair against each other displaced translationally, so that although a distinctness of the coils of a pair is achieved by the two coils are excited at different times by a moving away over it Fördergutteil. However, this translational shift also causes the benefits of using a coil pair to be lost again because the two coils of a pair are no longer electromagnetically equivalent. In addition, the gain in the spatial resolution is only about a factor of 2 and still depends on the size of the sensor coils as in the other prior art.

It is therefore the object of the present invention to provide a sensor device with sensors of higher resolution in a structurally simple and cost-effective manner, which reliably detects parts of various sizes. Furthermore, a sorting device is to be made available, with the better and more reliable sorting results can be achieved.

This object is achieved with a sensor device having the features of claim 1 and with a sorting device according to claim 17.

Thereafter, the sensor device according to the invention comprises a plurality of sensors, the detector coil pairs are arranged so that their standing parallel to the conveying plane cross-sectional areas against each other shifted center of gravity, and the connecting line between the centroids is oblique to the direction of movement of the transported material. With essentially mirror-symmetrical design of the detector coil pair, the inductive coupling to the alternating field is largely identical for both detector coils. Both coils are exposed to the same total magnetic flux, provided that the alternating field is approximately homogeneous from the point of view of the detector coils. Therefore, it will become in the case of the suitably chosen, that is to say, the difference taking into account the signs of the detector coil voltages automatically approximates a zero balance.

According to the claim, each sensor has two detector coils, which are advantageously wound in a D-shape and spaced from one another. The two coils of a pair can e.g. to be wound in opposite directions, but an adjustment can also be made with the measuring amplifier.

The invention only assumes that the two detector coils of a pair differ in the e.g. of a metal part disturbed alternating field react. Furthermore, the sensor signal should also make the location of the crossing of the sensor detectable. This is achieved by the cross-sectional areas of the detector coils of a pair in a plane parallel to the conveying plane of the conveyed material at least partially do not overlap by the centroid of these cross-sectional areas are shifted from each other. Furthermore, the connecting line between these centroids and the plane of symmetry of the coil pair should be inclined to the direction of movement. This symmetry break can be used to distinguish from the sensor signal which is easy to evaluate, e.g. a metal part to the left or right of the sensor center crosses the sensor. If, for example, the connecting line were parallel to the direction of movement, a part crossing the sensor to the left of the sensor center would not be able to be distinguished from a part crossing the sensor to the right of the sensor center.

For illustration, imagine two identical circular planar coils as a detector coil pair whose area vectors are oriented perpendicular to the conveying plane. According to the invention, the two circular area centers of the detector coils are spaced apart from each other and the connecting line is obliquely to the direction of movement, so that the two coils do not lie one above the other in their circular cross-sectional areas, but are arranged offset to one another both in the direction of movement and transversely thereto.

It is achieved with the embodiment according to the invention with advantage that a part that changes the alternating field due to its material property, so for example. forming a secondary alternating magnetic field in response to the alternating field, or e.g. has a permanent magnetic field from home, and which is moved past the sensor, first in time sequence, the direction of movement further upstream detector coil influenced. In this coil, a voltage which does not lift away with suitable subtraction is first induced. Only then does the detector coil arranged in the direction of movement come into the influence of the field inhomogeneity. The suitably formed difference signal of the detector coil pair reflects this information.

The formation of the conveyor is largely arbitrary within the scope of the invention. The sensor device may e.g. also be arranged at a drop distance, which is e.g. a conveyor such as a conveyor belt or a chute connects. The direction of movement and the orientation of the conveyor plane change in this special case on the fall path of Fördergutteiles. Other conceivable delivery devices are known in the art.

If the sensor in the undisturbed alternating field should show a difference signal significantly different from zero, eg a zero balance can be produced electronically in order to optimize the sensitivity. Even inhomogeneities of the undisturbed alternating field could thus be matched, albeit with great effort. Preferably, before the evaluation, first of all, a total signal is generated in pairs from the two detector coil voltages, the negative and positive values can accept. For example, a measuring amplifier of conventional design is used. The evaluation then takes place on the total signal of the sensor thus generated.

It can e.g. from the time length of the positive, i. with phase, and negative, i. antiphase, signal components of the total signal and the evt11. intermediate residence time are largely closed on the time required by the detected part to cross the sensor or the individual detector coils. This time is greater with a central crossing than with crossing at the edge of the sensor. Due to the inclination can still be clearly distinguished whether the part left or right of the sensor center has crossed the sensor. In addition to the temporal length of the positive and negative signal component, which when tilted, e.g. may be different, the location of the zero crossing also depends on the location of the crossing above the sensor. From this it can be calculated at which point the sensor was crossed.

According to the invention, therefore, the location of the passage can be determined very accurately from the evaluation of the throughput times of a conveyed material part for one and / or the other of the detector coils of a pair, possibly linked to the position of the zero crossing and the knowledge of the inclination. It is thereby a resolution below the width of the sensor or the coil pair possible. The resolution is no longer determined by the sensor or coil size, but essentially by the accuracy of the skew, the accuracy of the measurement of the sensor signals and the accuracy of the evaluation of the time course. In this way, with sensors of otherwise the same size, a multiple higher spatial resolution can be achieved, and it can, for example, several distributed to the sensor width and the sensor locally associated exhaust nozzles are controlled by a location accurate information only a sensor.

Further advantageous embodiments are specified in the subclaims.

In particular, it is possible with the inventive sensors with appropriate design of the evaluation device to obtain an image of at least the contour of the part that traverses the sensor device. The sensors are particularly sensitive to the entry or exit of a part in or out of the sensor area, while, for example, when the sensor is completely covered, the sensor delivers substantially no signal deviating from zero. By tracking the development of the sensor signal over time and by using other known interpolation methods and plausibility checks, a detailed image of the detected part can be obtained. Advantageously, therefore, the features of claim 15 are proposed. The evaluation of the time profile of the signals of several adjacent sensors can provide a relatively detailed picture of the shape and size of the detected part. It can thus be reliably detected, for example, whether a large, flat object sweeps over the sensor device, or, for example, a long, thinner object. This has considerable advantages over the previously customary single-sensor-oriented evaluation. For example, based on this pictorial information, for example, a suitable or a plurality of suitably positioned discharge nozzles can be actuated, for example, not to blow an object at its edge, whereby the part would essentially only be rotated, but to act on the geometric center of gravity of the part. By suitably controllable design of the exhaust nozzles, for example, the Ausblasimpuls can be adapted to the part size, eg application of a strong Ausblasimpulses for large parts and a smaller Ausblasimpulses for small parts. It can thereby minimize the energy expenditure.

The sensors could e.g. be arranged in any distribution in the sensor device. But this is disadvantageous for the evaluation and the control of the associated separation device. Advantageously, the sensors of the sensor device are therefore arranged in a row according to claim 2, which is perpendicular to the direction of movement of the vorbegutgescheweg Fördergutstromes. As a result, the running time of the parts from the sensor to the effective range of the separating device is the same for all sensors in the line and the control of the separating device is simplified.

The alternating field could e.g. be generated by an excitation coil extending over all sensors. As a rule, however, there is then no spatially very homogeneous field, so that the voltages induced in the detector coils of a pair cancel out only inadequately. Furthermore, e.g. Fe materials or other magnetizable materials in a larger alternating field comprising multiple sensors will result in significant field line constrictions which will also produce signals on adjacent but not traded detector coil pairs. Cross-sensitivity becomes unacceptable for some applications. Advantageously, therefore, according to claim 3 each sensor associated with an excitation coil. In particular, with the further advantageous features of claims 4 and 5, an improvement of the field homogeneity over the sensor coils can be obtained.

The advantageous features of claim 6 serve the same purpose. When all coils are arranged in the same plane, possible interfering influences due to sensor coils of adjacent sensors are reduced.

The sensor device according to the invention could for example be composed of a suitable number of individual sensors. According to claim 7 but is provided with advantage that several sensors combined on a Feinleitei board are arranged. In the layout of the board, the coils can be designed very accurately, and also the production can be done with high precision. The use of boards allows accurate and rapid positioning of the sensors in the sensor device. The sensor device can be constructed, for example, from a single, all sensors supporting board. But it can also be composed of several smaller sensor boards. An advantage of using smaller sensor boards is that variable widths of the conveyed stream or conveyor can be covered by the addition of further sensor boards. For example, a sensor board may have a linear array of five sensors. With eight boards, which are housed in a common housing, for example, so a sensor device can be constructed with a sensor array of two lines, each with twenty sensors. Furthermore, the use of smaller sensor boards, for example, allows a modular structure, the individual boards could, for example, be operated and evaluated modularly. Another advantage of the boards arises from the ability to produce the coils, so both the exciter and the detector coils, in modern fine conductor technology and thus very high geometric accuracy.

Disturbing interaction between the coils of adjacent sensors can be further reduced with the advantageous features of claim 8. The excitation coils are operated in frequency and in phase, for example in a range between 5 kHz and 1 MHz. It is thereby ensured, for example, that no crosstalk between adjacent sensors is detected, among other things, because each sensor has largely identically acting sensor neighbors whose influence can essentially cancel out due to the pairwise arrangement of the detector coils in total. Less environmental interference means improved measurement accuracy and higher sensor sensitivity.

With the features of claim 11, a possible interference by adjacent sensors is further reduced. Each sensor sees an identical neighborhood to the left and right of it, possibly in front of or behind it. Point symmetry at each sensor center would be optimal from a mathematical point of view. The ideal symmetry comes close to the features of claim 11.

In order to ensure as complete a detection as possible over the entire width of the sensor device, e.g. of metal parts, the sensors could e.g. be arranged as close as possible. Advantageously, the sensors are arranged according to claim 9 in a plurality of staggered rows, e.g. standing on a gap. Multiple rows of sensors can also be used to advantage to check results of the sensors in the one row for errors by comparison with results of the sensors in a second row.

The range and sensitivity of the sensors can be increased by having the detector coils and / or the exciting coils in accordance with claim 10 having a core, e.g. a ferrite core or a core of other suitable material.

Critical may be the edge regions of the sensor device, because the outermost sensors have only on one side of a neighboring sensor, the disturbance can not cancel so for reasons of symmetry. With the advantageous features of claim 12, however, it is achieved that the sensor located on the outside in the line is nevertheless exposed to approximately the same proximity influences, because in addition a sensor-free exciter coil is arranged next to it. It would also be possible to incorporate a further sensor into this marginal exciter coil which, for example, is not used for evaluation. The interference of adjacent exciter coils is much higher than the neighboring ones Detector coils. The additional effort would be in so far in no favorable ratio to additionally achieved parasitic reduction.

It proves to be advantageous to choose the angle between the direction of movement of the conveyed material and the connecting line of the centroids or the plane of symmetry of the detector coils of a pair between 30 ° and 60 °, in particular to select the inclination with 45 °. If this angle was chosen to be very small, ie if the first detector coil is almost completely in front of the second detector coil in the direction of movement, it is difficult to distinguish whether the detected part has crossed the sensor to the left or right of the sensor center. This distinction is not possible if the plane of symmetry is exactly perpendicular to the direction of movement. For the other extreme case, namely that the plane of symmetry is parallel to the direction of movement, the sensor can detect large or centrally over the sensor moving Fördergutteilchen only very bad, because in both detector coils substantially equal voltages are induced, which can cancel to about zero. A local resolution below the sensor width succeeds at the preferred 45 ° in an optimal manner.

The frequency of the alternating field can be selected within wide limits. For example, a monofrequent field can be selected. Preferably, however, the device for generating an electromagnetic alternating field for generating a multi-frequency alternating field is formed. The alternating field then represents a superimposition of several fields of different frequencies. For example, several discrete frequencies or eg a frequency band can be used. The use of such a frequency-mixed alternating field ensures that regardless of Fördergutteilgröße and Fördergutteilmaterial always reliable detection takes place. As a basic rule, it can be stated that for the detection of small filigree parts, eg of wires, higher frequencies are advantageous, for example, between 150 and 500 kHz, while the detection of iron parts in particular succeeds preferably at lower frequencies, for example less than 20 kHz. To generate such frequency-mixed fields, the excitation coils are subjected to corresponding signals.

The advantages of the sorting device according to the invention according to claim 17 result from the advantageous features of the sensor device used. The features of claims 18 to 20 relate to further advantageous embodiments.

Fig. 1

eine Prinzipdarstellung eines Ausführungsbeispiels einer erfindungsgemäßen Sortiervorrichtung in Seitenansicht,

Fig. 2a, 2b

eine Prinzipskizze zur Funktionsweise eines Ausführungsbeispieles eines Sensors der erfindungsgemäßen Sensorvorrichtung (Figur 2a) mit mehreren Signalverläufen des vom Sensor gelieferten Signals ( Figur 2b),

Fig. 3

eine Prinzipdarstellung einer möglichen Schaltungsanordnung für das Ausführungsbeispiel eines Sensors nach Figur 2a, und

Fig. 4

einen Ausschnitt in Draufsicht auf ein Ausführungsbeispiel einer erfindungsgemäßen Sensorvorrichtung in prinzipienhafter Darstellung.

Below, the invention will be explained with reference to exemplary embodiments, which are shown schematically and in principle in the figures. The same reference numerals stand for the same parts. Show it:

Fig. 1

a schematic representation of an embodiment of a sorting device according to the invention in side view,

Fig. 2a, 2b

a schematic diagram of the operation of an embodiment of a sensor of the sensor device according to the invention ( FIG. 2a ) with several signal curves of the signal supplied by the sensor ( FIG. 2b )

Fig. 3

a schematic diagram of a possible circuit arrangement for the embodiment of a sensor according to FIG. 2a , and

Fig. 4

a detail in plan view of an embodiment of a sensor device according to the invention in principle representation.

Fig. 1 shows in a graphically greatly simplified form the basic structure of a sorting device 10 for sorting out a metallic fraction 15, 15 ', 15 "from a Fördergutstrom 13. A conveyor belt 12 which is fed in a manner not shown with conveyed material 13, for example via an upstream chute, which in turn is loaded, for example, by a conveyed material supply, transports conveyed material 13 at a uniform speed over a sensor device 14 arranged below the belt 12. Details of the sensor device 14 are described with reference to FIG Figures 2-4 will be explained later.

The material to be conveyed 13 consists of a metallic fraction 15 and non-metallic Fördergutteilen 16. In the example shown, the individual parts of the conveyed significant differences in size. In alternative embodiments, e.g. be preceded by a screening step or the conveyed already present by a suitable treatment already in a uniform size.

The sensor device 14 is connected via a data bus 18 to an evaluation and control device 20. The object of this evaluation and control device 20 is to evaluate the sensor data supplied by the sensor device 14 to determine whether a part to be sorted out passes through the sensor region detected by the sensor device. Furthermore, it is then appropriate time-delayed to control the separator 22, so that a detected metal part 15 'is sorted out. The evaluation and control device 20 may e.g. also be integrated into the sensor device 14.

In the example shown, the separating device 22 consists of an exhaust nozzle 24, which is arranged below the conveyor belt 12 to a drop distance. From the conveyor belt 12 falling Fördergutteile 15 'can be blown from the exhaust nozzle 24 to a transversely accelerating pulse on the Fördergutteil exercise and distract it from the undisturbed trajectory to another, eg wider flight parabola. The exhaust nozzle 24 is dominated by a valve 26, for example by a solenoid valve. The control of the valve via control lines 28 of the evaluation and control device 20. The valve 26 is a compressed air hose 32 dominantly formed, which compressed air from a compressed air reservoir 34 leads to the exhaust nozzle 24. By timely and short-term opening of the valve 26 a vorbeifallendes on the nozzle 24 Fördergutteilchen 15 'is subjected to a Blasimpuls so that it is deflected by its undisturbed fall track and flying over a separating edge 36 to be transported away by another conveyor belt 40, eg to a further processing or to another sorting stage. Only rejected metal parts 15 "are to be found on this transporting conveyor belt 40. For example, the discharge nozzles 24 can be designed and controlled in such a way that the intensity of the blowout pulses can be selected to suit the parts to be sorted out.

Not to be sorted conveyed material parts 13 'fall undisturbed on a third conveyor belt 42, which this reduced by the metal fraction Fördergutfraktion. also transported to another processing or sorting.

The conveyor belt 12 has a certain conveying width, and the conveyed material parts 13 are moved across the width across the sensor device 14. Therefore, the sensor device 14 extends across the width of the conveyor belt 12. Within the sensor device 14, a plurality of sensors 100 are arranged distributed over its width, so that the width position of a metal part 15 on the conveyor belt 12 can be determined. Accordingly, the separator 22 has a plurality of arranged in a row transversely to the direction of fall discharge nozzles 24, which are arranged covering the Fallwegbreite suitably.

The evaluation and control device 20 is designed to control that or those multiple exhaust nozzles 24, which are assigned to the position of the sensor or the sensors in the sensor device 14, which have detected a conveyed item 15. Furthermore, the evaluation and sensor device 20 takes into account the transit time of a particle from the sensor arrangement 14 to the blow-off position, that is to say until the effective range of the blow-off nozzles 24 is reached. a communicating with the evaluation and control device 20 connected measuring device for detecting the belt speed may be provided. In the embodiment shown, an angle encoder 29 is arranged on the guide roller 27 of the conveyor belt 12, which measures the instantaneous speed of the guide roller, from which results in the conveyor belt speed. The evaluation and control device 20 calculates with this instantaneous speed the correct time for the triggering of the exhaust nozzle 24th

There are further alternative embodiments of this embodiment of a sorting device 10 conceivable. For example, the conveyor belts 12, 40 and 42 could be replaced individually or all by means of transport chutes or other conveying means, instead of the conveyor belts 40 and 42, containers could also be provided. The exhaust nozzles 24 could also be located above a drop path to effect a deflection of the sorted fraction 15 ', 15 "downwards from the free fall path The separation edge 36 would then be placed in a different position, but could be eliminated altogether, though Furthermore, it is possible to provide additional sensors, for example for observing the drop distance of the material to be conveyed, in order, for example, additionally to optically detect, for example, the metal parts 13 'detected by the sensor arrangement 14, in order to time the outlet nozzles 24 to drive even more precisely.

Fig. 2a and 2b show in a schematic representation of the operation of an excitation coil 102 and two oppositely wound detector coils 104, 106 existing sensor 100. In the example shown, all coils 102, 104, 106, two windings. However, the number of turns can be chosen differently, wherein the detector coils 104 and 106 should have the same number of turns. Not shown are the electrical lines through which these coils 102, 104, 106 are energized. The contact surfaces associated with the coils 102, 104, 106 are designated by reference numerals 112, 114 and 116.

The two detector coils 104 and 106 are concentrically surrounded as a coil pair of the exciter coil 102, which serves to generate an alternating field. By this alternating field magnetic fields are induced in metal parts 15A, 15B, 15C, which enter the effective range of the alternating field and the sensor 100. The exciter coil 102 is charged with a high-frequency alternating voltage, so it generates an electromagnetic alternating field with the same frequency. Typical frequencies can be e.g. in the kHz range. Frequency mixtures can also be used.

Material items 13 made of a non-conductive material show no interaction with the alternating field. In contrast, an electromagnetic alternating field is induced in typical conductive material 15A, 15B and 15C typical, material-equivalent transfer function.

The two detector coils 104 and 106 are arranged mirror-symmetrically to a mirror plane 115. This mirror plane 115 is inclined to the direction of movement 116 of the conveyed items 15A, 15B, 15C. In the example shown, the angle enclosed between the direction of movement 116 and the mirror plane 115 is 45 °. The centroid M of the detector coil 104 and the area center of gravity M 'of the detector coil 106 are connected by an imaginary connecting line L, which is oriented perpendicular to the mirror plane due to the 45 ° inclination.

In Fig. 2b three waveforms A, B, C of the signal supplied by the sensor 100 to the three metal parts 15A, 15B, 15C are shown, these three parts pass through the sensor 100 at different locations. Part 15A traverses the sensor 100 in the center, while the parts 15B and 15C cross the sensor 100 farther out, in the case of part 15C only at the outer edge.

If one follows the movement path 116 of the part 15A, it first traverses the detector coil 104 located further in the direction of movement 116. As can be seen from the upper signal curve A of FIG Fig. 2b As can be seen, this leads to an increase in signal, that is to say that the differential voltage between the two detector coils 104 and 106 which can be measured in phase increases because the detector coil 106 located further back in the direction of movement 116 is not yet influenced by the alternating field induced in metal part 15A. Without limiting the generality, let it be assumed that the detector coil 104 delivers a positive signal component, while the detector coil 106 supplies a negative signal component. Under this assumption, the total signal A increases as soon as the metal part 15A penetrates into the detection area of the detector coil 104. The signal A then reaches a constant value and then drops after a certain time corresponding to the duration of the crossing of the detector coil 104, to after a zero crossing, which corresponds to the crossing of the plane of symmetry 115 through the metal part 15A, in the negative region to go. Here dominates the negative signal of the detector coil 106. After leaving the effective range of the detector coil 106, the signal A falls back to zero.

The waveforms B and C shown in the metal parts 15B and 15C are the same ones. Metal part 15B reaches the area of influence of detector coil 104 a little later than part 15A. The traverse time for detector coil 104 is also shorter than for metal part 15A, and symmetry plane 115 is reached sooner so that the zero crossing also occurs earlier in time. The negative part of the total signal B is longer in time because the metal part 15B has to travel a longer distance across the detector coil 106. After metal particles 15B has also completely crossed over the second detector coil 106, here too the signal B drops back to zero, wherein metal part 15B leaves the sensor region in a shorter time than metal part 15A. Therefore, signal B is also shorter in time than signal A.

For metal part 15C, the special case is that the detector coil 104 is not crossed at all. Therefore, the signal curve labeled B does not show a positive total signal component. It is therefore missing at a zero crossing. As metal part 15C reaches detector coil 106, the total signal goes into the negative signal region. The metal part 15C leaves after a relatively short time detector coil 106, so that the negative total signal duration is shorter than in the waveforms A and B shown to the metal parts 15A and 15B.

The dash-dot lines shown parallel to the direction of movement 116 correspond to positions of the exhaust nozzles 24 associated with the sensor 100, which in FIG Fig. 1 were shown. Distributed over the width of the sensor 100 are seven of these blow-off nozzles 24, wherein the two outermost lying lying partially associated with this sensor 100, but in part also the left or right neighbor sensor.

Fig. 3 shows in a block diagram a possible wiring of a sensor 100, the circuit having only exemplary character and in particular at Use of multiple sensors 100 of which looks differently designed. For example, ADC chains or multiplexers, a row logic and parallel computers can be used. The architecture of the evaluation electronics is largely freely selectable and unaffected by the sensor structure. Such circuits are also generally known in the art, so it is not discussed further below.

The control could e.g. via a conventional computer with suitable interfaces for communication with the sensor 100, the angle sensor 27 and the exhaust nozzles 24 done. However, the data volumes generated by the sensors will be so significant that computers and interfaces will reach their performance limits. In the example shown here, the control, supply and signal evaluation of the sensor 100 is therefore taken over by an integrated powerful microcontroller (μC) 302. This microcontroller 302 can also take over the control of the exhaust nozzles 24, we is about an interface and a bus line 303 with the valves to be switched 26 in conjunction. But a microcontroller solution is only one of several possibilities.

The excitation coil 102 is acted upon by a power amplifier 305 with a suitable high-frequency AC voltage to produce an alternating field. It can also be given a frequency-mixed AC voltage to the exciter coil. The power amplifier 305 in turn is supplied by an upstream digital-to-analog converter (DAC) 307, which in turn is controlled by the microcontroller 302.

On the signal side, the AC signal which can be tapped off at the detector coils 104 and 106 is fed to a measuring amplifier 309, which is designed as a differential amplifier, and an analog alternating signal to an analog-to-digital converter (ADC) 311 provides, via which the signal of the sense amplifier 309 in turn gets back to the microcontroller 302. There or in another computer device, the signal is eg as above to Fig. 2b described evaluated.

The tasks of the microcontroller 302 could also be extended to the evaluation and overall control of the sorter 10 by suitable design, and may be e.g. be integrated into the sensor device 14 as digital hardware and firmware. Because of the parallel requirement of the same real-time mathematics for the multiple sensors 100 and the data streams generated thereby, the use of parallel computing and digital signal processors is beneficial.

Fig. 4 shows in plan view a section of a sensor device 14 with two rows of sensors 100, the the in Fig. 2a correspond to the embodiment shown. In this case, these sensors 100 are arranged in groups on a circuit board 402. Without limitation of generality, for example, five sensors 100 per board 402 may be provided in each case.

The sensors 100 are within the rows at equal intervals, the sensors 100 of the one line are in gap to the sensors 100 of the other line. All sensors 100 show the same inclination of the plane of symmetry 115 relative to the direction of movement 116, which here corresponds to the direction of the dot-dash lines, which, as already in Fig. 2a , the local arrangement of the exhaust nozzles V1-V20 (24) represent, which are driven in response to the sensor signals for blowing a detected metal part 15.

The distances between the sensors 100 within a row are smaller than the diameter of the detector coils 104, 106, so that due to the offset Arrangement of the two rows results in an overlap in width. A metal part, for example running along the path of movement that can be associated with the blow-off valve V5, passes over both sensor 100 'and sensor 100 " whether the detection of a metal part by sensor 100 'at a certain location coincides with the time-delayed detection message of the sensor 100 ". Furthermore, in such an arrangement, a plausibility check can be performed on parts that are wider than the valve spacing. For example, a part that covers the valve line V1 and V2, can be calculated in its contour: the first sensor 100 of the upper line responds with the same length positive and negative signal time. The sweeping at the level of the line V2 could not be measured by the first sensor 100. Since, however, the first sensor 100 "of the lower line now reacts exclusively with a positive signal, the part along the lines V1 and V2 must have crossed the sensors 100 and 100". If the part is even larger, so it sweeps over the sensors 100 and 100 "eg along the lines V1, V2 and V3, so the first sensor 100" of the lower line reacts with a long positive signal and a short negative. It can therefore be concluded that the sensors have been swept over the width of the lines V1, V2 and V3. It can be interpolated according to this principle, almost every conceivable part width, which is larger than the sensors themselves.

The sensor device 10 according to the invention permits high-resolution locating of electromagnetically detectable parts which are moved over the sensor device 14. By suitable evaluation of the sensor signals, for example by the signals of several adjacent sensors or even all sensors and by adding interpolation according to the just described principle, a complete image of the detected parts can be obtained. The Sensor device 14 shown by way of example functions reliably for a very wide range of sizes of the parts to be detected, that is to say both for parts which are smaller than the diameter of sensors 100 and for parts which run over a plurality of sensors 100 at the same time.

The after Fig. 4 resulting improvement of the resolution over the generic state of the art by a factor of 5 is already achieved by simple means. When using more computing and evaluation power, and possibly also when using other sensors, the resolution can be further refined without much effort. Resolution increases by a factor of 10 were easily achieved.

Claims (20)

1. A sensor assembly (14) comprising a plurality of sensors (100) for detecting electromagnetically detectable pieces (15, 16) of material to be conveyed, comprising an allocated conveying device (12) that moves material to be conveyed (13) past the sensors (100) in a conveying plane and in a direction of movement (116), comprising a device (102) for generating an electromagnetic alternating field, wherein the sensors (100) each include a pair of detector coils (104, 106) which are connected to an evaluation device (20) for determining a differential signal between the coils (104, 106) of a pair in order to detect pieces (15) of material to be conveyed which deform the alternating field on the basis of their material properties, wherein the two detector coils (104, 106) of the pair are arranged such that they comprise cross-sectional areas extending in parallel to the conveying plane and comprising centroids (M, M') that are displaced in relation to each other, wherein the two detector coils (104, 106) of the pair are formed in the shape of a D and, being spaced apart from each other, are arranged mirror-symmetrically in relation to a plane of symmetry (115) which is not simultaneously a mirror plane of symmetry of each of the detector coils (104, 106), wherein the connecting line (L) between the centroids (M, M') and the plane of symmetry (115) is arranged at an angle in relation to the direction of movement (116) of the material to be conveyed (13).
2. The sensor assembly (14) according to Claim 1, characterized in that the sensors (100) are arranged in a line which, in essence, extends perpendicularly to the direction of movement (116) of the material to be conveyed (13).
3. The sensor assembly (14) according to anyone of the preceding claims, characterized in that the device for generating the alternating field comprises a plurality of excitation coils (102), with one excitation coil (106) being allocated to each sensor (100).
4. The sensor assembly (14) according to Claim 3, characterized in that the sensor coils (104, 106) are surrounded by the allocated excitation coil (102).
5. The sensor assembly (14) according to anyone of Claims 3 or 4, characterized in that the excitation coils (102) are circularly wound coils.
6. The sensor assembly (14) according to anyone of the preceding claims, characterized in that all of the excitation and detector coils (102, 104, 106) are, in essence, arranged in the same plane extending in parallel to the conveying plane of the flow of material to be conveyed (13) passing by.
7. The sensor assembly (14) according to Claim 6, characterized in that a plurality of sensors (100) is arranged on a fine printed circuit board (402).
8. The sensor assembly (14) according to anyone of the preceding claims, characterized in that all of the excitation coils (102) are operated synchronously.
9. The sensor assembly (14) according to anyone of the preceding claims, characterized in that the sensors (100', 100") are arranged in a plurality of lines that are offset in relation to each other.
10. The sensor assembly (14) according to anyone of the preceding claims, characterized in that the detector coils (104, 106) and/or the excitation coils (102) include a core that is open towards the flow of material to be conveyed (13).
11. The sensor assembly (14) according to anyone of the preceding claims, characterized in that the sensors (100) positioned in a line are arranged such that they are equally spaced apart from each other.
12. The sensor assembly (14) according to anyone of the preceding claims, characterized in that the excitation coils (102) arranged at the ends of the line do not comprise any sensors.
13. The sensor assembly (14) according to anyone of the preceding claims, characterized in that the angle between the direction of movement (116) of the flow of material to be conveyed (13) and the connecting line of the centroids and/or the plane of symmetry (115) of the detector coils (104, 106) of a pair ranges from 30 degrees to 60 degrees.
14. The sensor assembly (14) according to Claim 13, characterized in that the angle is 45 degrees.
15. The sensor assembly (14) according to anyone of the preceding claims, characterized in that the evaluation device (20) is designed so as to evaluate in a chronologically correlated manner the signals of a plurality of sensors (100) that are arranged adjacent to each other.
16. The sensor assembly according to anyone of the preceding claims, characterized in that the device (102) for generating an electromagnetic alternating field is designed for the generation of a multi-frequency alternating field.
17. An apparatus (10) for automated sorting of electromagnetically detectable fractions out of a flow of material to be conveyed (13), comprising a conveying device (12) for moving the material to be conveyed (13) through a detection range, comprising a sensor assembly (14) arranged in said detection range which is provided for detecting electromagnetically detectable pieces (15) of material to be conveyed in a flow of material to be conveyed (13) that is moved in transverse direction in relation to and past said sensor assembly (14), comprising a separating device (24) for selectively separating a detected fraction (15', 15") to be sorted out, and comprising an evaluation and control device (20) activating the separating device (24) based on the evaluation of the results provided by the sensor assembly (14), characterized by a sensor assembly (14) according to anyone of the preceding claims.
18. The apparatus (10) according to Claim 17, characterized in that the separating device comprises a plurality of blow-off nozzles (24).
19. The apparatus according to Claim 17 or 18, characterized in that the conveying device is a conveyor belt (12) and that the sensors (100) are arranged underneath said conveyor belt (12).
20. The apparatus according to Claim 19, characterized in that a measuring device (29) determines the instantaneous transport velocity of the conveyor belt (12) and that this measurement is used for correcting the delay time between the detection of the piece of material to be conveyed (13) and the activation of the separating device (24).

Images