WO1994023850A1

WO1994023850A1 - Device for assessment of moving objects

Info

Publication number: WO1994023850A1
Application number: PCT/EP1994/000929
Authority: WO
Inventors: Hans P. Friedrich; Andreas Dux
Original assignee: Hpf Hans P. Friedrich Elektronik Gmbh
Priority date: 1993-04-17
Filing date: 1994-03-23
Publication date: 1994-10-27
Also published as: DE4312550C1

Abstract

A device for inspecting and/or differentiating falling objects (11, 12), preferably free-falling tablets (13), has an illuminating device designed to produce a first distribution of light intensity in a first measuring plane, influencing the given object (12, 13) as it traverses the first measuring plane, and to produce a second distribution of light intensity in a second measuring plane that is different from the first measuring plane, influencing the given object (12, 13) as it traverses the second measuring plane. A detection device is combined with the illuminating device to form a measuring station (18), the detection device being designed to measure in two successive measuring steps the influences of the first and second intensity distributions along first and, differing therefrom, second measuring coordinates in such a way that it detects the respective dimensions of the object (12, 13) in the direction of the respective measuring coordinate. Provided also is an analysis device (19) which derives from the detected dimensions at least one control signal (S1, S2), which comprises a qualification of the particular object (12, 13), such that a downstream sorting station (21) dispatches the objects (12, 13) differently depending on the control signal (S1, S2).

Description

Device for evaluating moving objects

The present invention relates to a device for checking and / or distinguishing falling objects, preferably free-falling tablets. The pharmaceutical industry is moving more and more to sell tablets no longer in blister packs, but again conventionally in tablet tubes. This poses the problem of how tablets are to be introduced into the bottles in terms of number and quality.

So far, this has been accomplished by pulling strips with troughs adapted to the tablets to be filled through a "swamp" of tablets, one tablet being deposited in each trough. These strips are then emptied via filling funnels, under which the tubes to be filled pass. However, this method has the disadvantage, on the one hand, that it is inaccurate and, on the other hand, that the tablets cannot be checked for breakage simultaneously during filling.

However, it is already known to use devices for counting and for breaking detection, using cameras and monitoring the tablets while they are being filled into the tubes. However, these devices have several disadvantages:

On the one hand, they need an image memory, which requires complex hardware. On the other hand, since the image memory must first be fully written before the evaluation can take place, these conventional systems are very slow, and their resolution is also limited. Finally, since all cameras require imaging optics, spherical aberrations have a distorting and accuracy-limiting effect on the evaluation. In addition, considerable space requirements are associated with camera systems. A device known from document DE-U-9114260 is used to check tablets in blister packs for breakage. For this purpose, the blister packs are moved past an LED line, which produces a kind of light curtain, before the lid film is applied. The tablets arranged side by side in several lanes disrupt its intensity distribution when passing through this light curtain, which is detected by a line sensor either in the incident light or in the transmitted light method.

The LED line and the line sensor are clocked in such a way that a series of sectional images of the tablets crossing the measuring plane formed by the light curtain are generated as the blister packs move past. Furthermore, the feed speed of the blister packs is detected, so that the line sensor at known intervals shows the dimensions of the tablets in the longitudinal direction of the line sensor - that is, in a measuring coordinate. From the temporal course of the intensity distribution measured in this way, the area of the tablet transversely to this measurement plane is then, as it were, an evaluation criterion. On the basis of this evaluation, a control signal is generated which indicates whether the tablet is error-free or not.

If the calculated area falls below a reference value, an error is output, ie the tablet is marked as broken or faulty. The reference value is determined by first reading a blister pack with "good" tablets. An infrared LED line is used to avoid extraneous light influences. Accordingly, an infrared filter is connected upstream of the line sensor. In order to obtain a sharp image of the individual sections through the tablets, imaging optics in the form of a row of cylindrical lenses - so-called SELFOC lenses - are arranged between the infrared filter and the line sensor.

With regard to the tablets sold in blister packs, the disadvantages of the methods mentioned at the outset are therefore already eliminated. For each cut, the line sensor is read out serially and a comparator is used on-line to decide whether or not the currently read pixel is to be assigned to a tablet. The pixels to be assigned to a tablet are fed to the respective counter for the corresponding column and counted there via a data selector. The number of all pixels that were determined for a particular tablet is then compared with a previously read reference value.

This arrangement therefore has the advantage that, on the one hand, the complex hardware of known camera systems can be dispensed with and, on the other hand, the evaluation is carried out on-line.

Another device for checking tablets inserted in blister packs is described in US Pat. No. 4,847,487. Every possible tablet position there is a light transmitter and a light receiver in the form of a phototransistor assigned. The geometrical arrangement is such that a complete tablet completely darkens the phototransistor. If the tablet has defects that are noticeable in their cross-section, depending on the size of this defect, more or less light reaches the phototransistor, which recognizes that a defective tablet has been inserted. On the basis of the intensity of the incident light, the phototransistor can also recognize whether an only partially defective tablet has been inserted or whether the tablet is completely missing. On the basis of this result, the known device outputs a control signal that rejects the entire blister pack checked, that is, excludes it from further processing.

Furthermore, it is known from utility model DE-U-73 21096 to use light barriers in packaging machines. The light barriers here control two-armed levers that get into the path of the light barrier in the absence of an object in the relevant path of the packaging machine.

Starting from this prior art, it is an object of the present invention to provide a device with which falling objects can be checked and qualified with a simple design effort. In particular, it should be possible to evaluate freely falling objects.

With regard to the device mentioned at the outset, this object is achieved in that the following features are provided: a lighting device which is set up to generate a first intensity distribution of light in a first measuring plane, which influences the respective object when passing through the first measuring plane, and to produce a second intensity distribution of light in a second measuring plane different from the first measuring plane , which influences the respective object when crossing the second measuring plane,

a detector device which is set up to measure the influencing of the first intensity distribution along a first measuring coordinate and the influencing of the second intensity distribution along a second measuring coordinate different from the first measuring coordinate in such a way that they measure the respective dimensions of the object in Direction of the respective measurement coordinates detected, and

an evaluation device which derives from the detected dimensions at least one control signal which includes a qualification of the respective object, wherein

a resettable sorting station forwards the objects differently depending on the control signal.

The object underlying the invention is completely achieved in this way. The fact that the influencing of different intensity distributions by the object is measured in two mutually different measuring coordinates can namely Obtain statements about the quality of the object falling through the measurement planes. The objects are assessed in two different views, so to speak, which significantly increases the possibility of checking them. While it was only possible in the known devices to check the "floor plan" of the objects, so to speak, so that errors that could only be recognized from the side were not taken into account, the objects were "rounded" here. With this device it is now also possible to check and distinguish objects whose exact orientation in space is not known.

In this way it is not only possible to examine tablets for breakage, it is also possible to differentiate between different types of tablets.

The evaluation of the respective object can either include a comparison with a reference object, but it can also be possible to determine the contour of the respective object, so that errors are also recognized that compensate for themselves in the area. That So that even a partially too small or partially too large area, which is not recognized as an error when the recorded area elements are added up, can still be assessed as defective here.

A further advantage is that, because of the sorting station that can be added, the objects can either be separated out or further processed depending on the result of the evaluation. A distinction between different objects can also be provided. It is thus possible, for example, to use the new device to simultaneously check different types of tablets and to forward them to different tubes or in different numbers to common tubes. It may be necessary, in particular, to ensure that the objects maintain their orientation transverse to their direction of movement at least when crossing the measurement planes. In this way it is excluded that tumbling objects lead to a falsified measurement result.

As before, the new device can be operated either by incident light or by transmitted light.

If the first and the second measuring coordinates lie in the first and second measuring plane, this has the advantage that the transmitted light method can be used directly, which is particularly desirable for adjustment reasons.

It is further preferred if the first and the second measurement coordinates run approximately perpendicular to one another.

The advantage here is that the object is viewed from the front and from the side, so to speak, so that comprehensive information about its state can be obtained.

It is further preferred if the detector device comprises a line sensor with optical sensors arranged side by side in the direction of the respective measurement coordinate.

This measure has the advantage that while the object is passing through the measurement plane, all sensors simultaneously record the intensity distribution; it is not necessary to scan the measurement coordinate in a mechanical or other way. For example, a CCD line can be used here. Furthermore, it is preferred if the first and second measurement planes run parallel to one another or preferably lie in the same spatial plane.

The advantage here is that no geometric distortions or the like have to be calculated from the measured intensity distributions. Since the distances at which the sections are taken from the objects depend on the speed at which the objects pass through the respective measuring plane, there are particularly simple relationships when the measuring planes are parallel to one another, because then the speeds of passage are the same for both measuring planes.

If the two measurement planes lie in a spatial plane, this has the further advantage that the overall height of the new device in the direction of movement of the objects is kept very low. This is a further advantage over the known camera systems, which have larger dimensions. If, on the other hand, the measurement planes are somewhat offset from one another in the direction of movement of the objects, the advantage of the low overall height is not impaired as compared with conventional camera systems, while there is the further advantage that the intensity distributions do not interfere with the detectors of the other measurement plane. This is promoted by a right-angled arrangement of the line sensors, but an additional offset brings further "optical isolation" of the measurement planes from one another. It is further preferred if the detector device comprises an imaging device which images the intensity distribution on the sensors.

The advantage here is that the imaging device ensures a high luminous efficiency. In addition, the image leads to a very sharp cut through the respective object, so that distortions such as can occur in camera systems are avoided here.

In a preferred development, the imaging device comprises at least one imaging light guide arrangement.

With this measure, it is particularly advantageous that the detector device itself does not have to be brought directly to the possibly falling object. Rather, this can be done by the imaging light guide arrangements, which transmit the intensity distribution mapped onto them without distortion to the detector device. Since the detector devices are generally provided with further electronics, a very compact, small measuring arrangement can be constructed from the lighting device and the light guide arrangement, which can be brought close to the objects to be checked. This eliminates all distortion errors that are caused by large imaging optics.

It is further preferred if the illumination device has at least one infrared light source and if the imaging device comprises an infrared filter. This measure has the advantage already discussed above, namely that the influence of extraneous light is eliminated. It should be mentioned here that these advantages can also be achieved with other optical wavelengths. It only has to be ensured that there is little or no extraneous light in the area of the new device at the wavelength used.

It is further preferred if the lighting device comprises light transmitters arranged in the form of cells in the respective measurement plane.

It is advantageous here that the intensity distribution in the measuring plane is built up in a stationary manner, so to speak. This light transmitter can either be an LED line or also, if necessary, also an imaging light guide.

Furthermore, it is preferred here if a device is provided which is set up to specify or determine the speed at which the objects move in the area of the measurement planes.

This measure has the advantage that influences on the speed of movement of the objects can be eliminated, so to speak. This is also advantageous in the case of freely falling objects, since the falling speed of objects of different weights differs from one another because of the disturbing influences of the air resistance. If this effect were not taken into account, fewer sectional images per unit of length would be taken from objects falling faster than from objects falling more slowly. On the other hand, if the analysis is nevertheless based on a constant spacing of the sectional images on the objects, faster falling objects would occur Objects, so to speak, measured as "too small". In a further development it is preferred if the device is assigned to a counting station which uses the control signal as a counting pulse.

The advantage here is that the control signal, on the basis of which "good" and "bad" objects are distinguished, is used at the same time to count the good objects. In this way e.g. the number of tablets can be recorded which have already been inserted into a tube to be filled.

Overall, it is preferred if the evaluation device carries out at least one comparison with at least one reference signal and generates the control signal as a function of the comparison.

The advantage here is that no complex and, so to speak, absolute evaluation criteria are required. In the form of a YES / NO decision, it can rather be decided in a simple and quick manner whether the object to be examined fulfills the evaluation criterion or not.

Overall, it is preferred if the device further comprises a release device for the defined and individual release of the objects, this release device being arranged upstream of the measurement planes in the direction of the objects.

The advantage here is that the objects are released one after the other to the measurement planes, and the objects can be given a known speed by the defined release. This has the further advantage that the above-mentioned arrangement for determining the speed at which the objects move in the area of the measuring planes, can be dispensed with. The arrangement mentioned and the release device mentioned here can thus be used alternatively, so to speak, in order to increase the measuring accuracy.

Furthermore, the release device has the advantage that the falling objects are separated, that is to say they penetrate the measurement planes one after the other. The release device can be set so that there is a certain time interval between two objects, so that the evaluation of the measurement of the first object can be concluded with sufficient certainty.

It is advantageous if the release device comprises at least one friction wheel which is resiliently mounted transversely to the fall direction.

This measure is structurally advantageous, because the falling object hits the friction wheel, deflects it to the side against the spring force and then moves between e.g. a counter wall and the resilient friction roller further down. The friction roller decelerates the falling object, the strength of the spring force exerting a direct influence on the falling speed of the object when it is released again by the friction roller.

It is also preferred here if the friction wheel is driven.

This measure allows speeds to be given to the released object which clearly deviate from the falling speed. The object can e.g. either accelerated or decelerated strongly, depending on the requirements of the evaluation in the detector device. It is further preferred if two friction wheels are provided.

Experiments have shown that, in particular, two friction wheels, between which the object moves, are advantageous, since the object is accelerated or braked evenly from both sides, so to speak, so that the object does not jam or spin when falling further.

It is further preferred if the device comprises an arrangement which determines or measures the speed at which the objects move in the area of the measurement planes.

This arrangement can either be a separate arrangement, e.g. The falling speed of the objects is determined or measured via light barriers, but it can also be implemented implicitly, so to speak, by the two measuring planes if these are arranged parallel to one another at a known distance. In the latter case, the time that the falling object takes to get from the first to the second measuring plane would be measured. The movement speed of the object could then be determined from the measured time and the known distance.

Further advantages result from the description and the attached drawing.

It goes without saying that the features mentioned above and those yet to be explained below can be used not only in the respectively specified combinations, but also in other combinations or on their own without departing from the scope of the present invention. An embodiment of the invention is shown in the drawing and is explained in more detail in the following description. Show it:

1 shows a filling station for tablets, in which the new device is used;

FIG. 2 shows a plan view of the measuring station of the new device, along the line II-II from FIG. 1;

3 successively measured intensity distributions for a falling tablet, measured in two mutually perpendicular measurement coordinates;

4 shows a side view of the measuring station of the new device, along the line IV-IV from FIG. 2;

5 shows a schematically indicated arrangement for determining the falling speed of the tablets from FIG. 1;

FIG. 6 shows, in a representation like FIG. 2, a measuring station in which the imaging device uses optical fibers;

FIG. 7 shows a block diagram of the evaluation device for the new device from FIG. 1; and

8 shows a schematically indicated device for the defined release of falling objects. A filling station 10 for objects 12, which in this case are tablets 13, is shown schematically in a side view in FIG. 1. The tablets 13 are placed in a chute 14, where they move downwards in the direction of movement 15, that is to say fall.

On their way through the chute 14, the tablets 13 pass through a device 17 for evaluating the objects. The device 17 comprises a measuring station 18 to be explained in more detail as well as an evaluation device 19 which is connected to the measuring station via lines 20.

A sorting station 21 is also provided, which either directs the incoming tablets 13 via a shaft 22 into a collecting container 23 for broken tablets 13a, 13b, or else fills the tablets 13c into tablet tubes 25a, 25b, 25c via a filling shaft 24. In the exemplary embodiment shown, the tablet tube 25a is already completely filled with tablets, while the tablet tube 25b is being filled.

The sorting station 21 comprises a flap 27 which is actuated by a control signal S1 which is generated by the evaluation station 19 and contains an evaluation of the objects 12. Depending on the position of the flap 27, the tablets 13 either get into the collecting container 23 or into a tablet tube 25. The evaluation device 19 outputs a second control signal S2, which reproduces the number of tablets 13 guided into the filling shaft 24. This control signal S2 is used to move the tablet tubes 25 further in the direction of their transport direction 28 when the predetermined number of tablets has been entered into the current tablet tube. The device 17 thus serves on the one hand to sort out defective tablets 13 or tablets 13 that are not to be used for other reasons. On the other hand, it also serves to fill a predetermined number of "good" tablets into each tablet tube 25.

The filling station 10 can be expanded in such a way that further sorting stations 21 are connected downstream of the filling shaft 24, so that different types of tablets can be checked and evaluated simultaneously by the device 17. In addition to the GOOD / BAD decision, the assessment then includes a further qualification with regard to the tablet type. In this way it is also possible to fill 25 tablets of different types into the same tablet tube.

FIG. 2 shows a plan view along the line II-II from FIG. 1 of the measuring station 18.

It can be seen that the measuring station 18 comprises an illuminating device 31 and a detector device 32, in the center of which a tablet 13 is indicated schematically. The lighting device 31 comprises a first infrared light source 33 in the form of an LED line 34, and a second "infrared light source 35 in the form of an LED line 36. The LED line 34 generates an intensity distribution of infrared light in a first measurement plane 37, while the LED line 36 produces a corresponding intensity distribution in a second measurement plane 38. The directions of propagation of the intensity distributions emitted by the LED lines 34 and 36 are indicated by arrows 39 and 40. It can be seen that the directions of propagation 39 and 40 are approximately are perpendicular to each other. A first line sensor 41 lies in the measurement plane 37, while a second line sensor 42 lies in the measurement plane 38. Each line sensor 41, 42 has side-by-side optical sensors which respond to light of the wavelength as emitted by the LED lines 34 and 36.

The intensity distributions in the measurement planes 37 and 38 are imaged on the line sensors 41 and 42 via imaging devices 43 and 44. These imaging devices each include an infrared filter 45 and an array 46 of cylindrical lenses, so-called SELFOC lenses.

The arrangement is such that the light emitted by the LED lines 34 and 36 is focused directly on the line sensors 41 and 42, the light of the first measuring plane

37 does not affect the second line sensor 42 and vice versa. This is due to the right-angled arrangement. If the measurement planes are in spatially offset planes, this optical isolation is further reinforced.

If a tablet 13 now passes across these measurement planes 37 or

38, this leads to a kind of shadow formation on the line sensors 41 and 42. As indicated in a coordinate cross 47, the dimensions of the tablet 13 are then in a first measurement coordinate 48 (line sensor 41) and a perpendicular second measurement coordinate 49 (line sensor 42 ) detected. The intensity distributions detected by the line sensors 41, 42 are fed via lines 20 as II and 12 into the evaluation device 19. In Fig. 3, two timing diagrams 51 are shown, which the. reproduce chronologically successive intensity distributions when a tablet 13 passes through the measurement planes 37 and 38. To this end, it should first be noted that both the LED lines 34 and 36 and the line sensors 41 and 42 are clocked, that is to say they are controlled at a certain time interval in such a way that they build up and query an intensity distribution. In this way there are successive cuts through a tablet 13 falling through the measurement planes 37, 38. These intensity distributions 53 are therefore at a distance 54 from one another which, together with the falling speed of the tablet 13, determines the spatial position of the individual cuts determined on the tablet 13. As can be seen from the time diagrams 51 and 52, the intensity distributions 53 lined up next to one another represent the view of the tablet 13 in the direction of the respective direction of propagation 39, 40. The dimensions of the tablet 13 can be calculated from these intensity distributions 53 in the direction of the respective measurement coordinate 48, 49 by counting the individual sensors of the line sensor 41 and 42, respectively, which are darkened by the tablet 13. In the direction of movement of the tablet 13, however, ie in the direction of the time axis t, the dimensions are determined from the time intervals 54 and from the falling speed of the respective tablet. If one knows the distance corresponding to the time interval 54, which the tablet has covered in this time interval, the time axis t can be transformed into a route axis and thus the area of the tablet 13 viewed from the direction of the propagation directions 39 in the manner of a support point integration. Calculate 40. Alternatively, it can also be provided that the contour of the tablet, as can be seen from the “silhouette” in FIG. 3, is evaluated directly. It could be the case that errors in the contour of the tablet are just averaging out with respect to the area of the tablet, but these errors should nevertheless be recorded.

Instead of a resolution of the tablet in the direction of the measurement coordinates, it is also possible to use a sum sensor in each measurement plane 37, 38, which measures the respective intensity distribution according to their sum. The shadow formation of a falling tablet 13 would then make this sum signal smaller, which can also be evaluated as a measure of the darkened area in the intensity distribution.

If the falling speed of the tablet 13 is always constant, no further measures are necessary, the time axis t is converted into a distance using a constant.

However, if tablets of different weights are to be sorted and evaluated, a different falling speed may result due to the disruptive influence of the air resistance.

FIG. 4 shows a sectional illustration along the line IV-IV from FIG. 2, by means of which a possibility for determining the falling speed is illustrated.

4 that the first measuring plane 37 and the second measuring plane 38 are offset from one another in the direction of the movement 15 of the tablet 13. Although the two measurement planes 37 and 38 run parallel to one another, they are at a distance 55 from one another, which makes it possible to determine the falling speed to determine. The time difference that lies between the immersion of the tablet 13 in the measuring plane 38 and the immersion in the measuring plane 37 determines the respective falling speed via the distance 55.

5 shows a further arrangement 57, which determines the falling speed of the tablet 13 with the aid of two light barriers 58 and 59 connected upstream or downstream in the direction of fall of the measuring planes 37, 38. The light barriers are arranged at a distance 60 from one another in the direction of movement 15 and each have a transmitter diode 61 and a receiver diode 62. The time between the interruption of the first light barrier 58 and the second light barrier 58 here determines the falling speed via the distance 60. Here, the two measurement planes 37, 38 can then lie in the same spatial plane, so that it is ensured that the passage speed is identical for both measurement planes 37, 38.

In FIG. 6, in an arrangement like FIG. 2, a measuring station 18 is shown, in which the imaging devices 43, 44 comprise optical fibers 64, which are arranged between the arrays 46 and a line sensor 66. These light guides transport the recorded intensity distribution without distortion to the line sensor 66, which in the exemplary embodiment shown consists of five individual segments 67. Instead of a line sensor 66, it is of course also possible to use an area sensor which is suitably illuminated by the light guides.

The use of such imaging light guides has the advantage that the intensity distribution can in each case be measured very close to the falling object, so that distortions are eliminated as far as possible. In addition, the Another advantage is that line sensors 66 can be used, the length (ie the number of sensors times the mean distance of the sensors from one another) is not adapted to the measuring field (the individual measuring levels). The imaging light guides 64 also make it possible to effectively use such line sensors 66. In the example shown in FIG. 6, the line sensor 66 has five segments 67, as mentioned, of which, however, only two and a half segments are required for a chute 14. The other two and a half segments can be used for another chute. It should also be noted here that the assignment of segments and imaging light guides in FIG. 6 is chosen only as an example.

Each of these segments 67 outputs its own intensity distribution J1 ^* , J2 ^* , J3 ^* , J4 ", J5 ^* , but these intensity distributions J ^* do not correspond to the actually measured intensity distributions J1 and J2. Like the conversion or reshaping 7 will now be explained with reference to the block diagram.

As already mentioned, FIG. 7 shows a schematic block diagram of a part of the evaluation device 19 from FIG. 1.

The intensity distributions J1 ^* ... J5 ^* read out serially are put into a comparator 71 separately for each segment 67, where they are compared with a threshold value 72 (K). The threshold 72 is set so that disturbing noise effects are compensated for. The comparators 71 indicate at their exit in the form of a yes / no decision whether the pixel in question is to be assigned to a tablet 13 or not. All five comparators 71 feed into a multiplexer 73, which assigns the individual pixels to the actually measured intensity distributions - here, for example, J1. At its output J1, the multiplexer 73 outputs all the pixels of the segment J1 ^* and as many pixels of the segment J2 ^* as are covered by the light guide 64 depicting the left in FIG. 6. At a further output, not shown, the multiplexer 73 outputs all pixels that belong to the intensity distribution J2 and come from the segments J2 ^* and J3 ^* .

The intensity distribution J1 is also fed serially, ie pixel by pixel, to a counter 74 which counts up the number of pixels assigned to a tablet 13 in an intensity distribution Jl. If a sum sensor is used, the counter 74 is of course unnecessary, since the sum sensor itself carries out the “summation”. If, on the other hand, the contour of the tablet 13 is to be evaluated, the counter 74 is likewise superfluous, because then the entire “length information” is evaluated.

After the number of pixels assigned to a tablet has been determined, this number is multiplied in a multiplier 75 by a factor F, which is supplied from a computing unit 76. The computing module 76 calculates the falling speed of the tablet 13 from its input signals 77, which originate, for example, from the receiver diodes 62 from FIG. 5, and outputs a corresponding factor on its output line 78. The resulting partial area value of the tablet 13, which corresponds to the intensity distribution J1 at the time, for example t_ (see FIG. 3), is stored in a summer 80. After a certain period of time, which can be determined by the falling speed, the next intensity is compiled and processed as described above. The area value of the tablet 13 corresponding to the temporally subsequent intensity distribution is added to the previously calculated area value in the summer 80. In a second adder 80 ', the area values of the tablet 13, which result from the measurement in the second measurement plane 38, are added up at the same time or with a time delay.

The outputs 81, 82 lead the summers 80, 80 'to a computing circuit 84 which combines the two summed values with one another and generates a comparison value V, which is fed to an comparator 86 at an output 85. In the comparator 86, the comparison value V is compared with a reference value R, which is supplied via an output 87 of a read memory 88.

This reference value R may either have been written into the read memory 88 from the outside, or may have been fed into the read memory 88 via a line 89 from the arithmetic circuit 84 during a “learning phase”. With this circuit it is possible to first drop a number of "good" tablets 13 through the chute 14 and to store the comparison values V determined in the process as "good values" in the read memory 88 for later use as a reference value. Exact knowledge of the geometry or the fall orientation of the objects is not necessary, the new device adapts itself automatically to the respective shape, so to speak. In the case of objects in particular, the orientation of which changes transversely to the direction of fall from object to object, a corresponding number of reference measurements ensures the required static safety for the reference value R. In the internal calculation of the signals in the arithmetic circuit 64, a reference signal supplied via a line 90 can also be taken into account, which contains a type of error evaluation criterion.

The comparator 86 now checks whether the comparison value V matches the reference value R within an error threshold Δ supplied via a line 91. The result of this comparison is displayed on an output line 92, which directly supplies the control signal S1 for the control of the sorting station 21.

The "good" tablets 13 are counted up in a downstream counting station 93, so that information is available at all times about how many tablets have already been inserted into a tablet tube 25b to be filled. This information is made available via an output line 94 as control signal S2.

If only one type of tablet 13 is measured at a time, the speed constant automatically results when teaching. The reference value R in the read memory 88, which is formed in exactly the same way as the comparison value V, then implicitly contains the falling speed, so to speak, so that the multiplier 75 and the arrangement 76 can be dispensed with. However, if you change the drop height of the tablet, it may be necessary to learn a new reference value.

The resolution that can be achieved with the device described so far is approximately 0.2 mm in the falling direction and approximately 65 μm transversely to the falling direction. While the resolution perpendicular to the fall direction is determined by the distance between the sensors, The resolution in the direction of the fall results on the one hand from the falling speed of the tablets 13 and on the other hand from the very high line repetition frequency of a line sensor compared to camera systems. The further advantage of a line sensor compared to a camera lies in addition to the higher spatial resolution in the line direction and the higher line repetition frequency, which means a higher resolution in the falling direction, and also in the smaller geometrical dimensions, which leads to a reduced overall height. In the case of tablets with a very small diameter in particular, the increased spatial resolution has great advantages over measuring devices working with camera systems.

FIG. 8 shows an alternative to the arrangement 57 from FIG. 5 or to the spatially spaced arrangement of the measuring planes 37, 38 from FIG. 4. FIG. 8 shows a schematic and partial side view of the device 10 from FIG. 1.

The tablets 13 falling in the chute 14 in the direction of movement or the direction of fall 15 reach the area of a release device 101 for the defined and individual release of the tablets 13 when they leave the upper part of the chute 14. The release device 101 is thus arranged upstream of the measuring station 18 in the direction of precipitation 15. After passing through the measuring station 18, the tablets 13 arrive in a second chute 14 ', which has larger cross sections than the first chute 14.

The release device 101 comprises two friction wheels 102, 103, which are arranged diametrically opposite one another and are biased toward one another via compression springs 104, 105. The two friction wheels 102 and 103 are driven in opposite directions, as indicated by arrows 106 and 107. The friction wheel 102 rotates counterclockwise and the friction wheel 103 clockwise.

8 that the tablets upstream of the release device 101 form a stack of tablets 13, 13a and 13b which is held by the release device 101. Further tablets 13 falling downward in FIG. 8 are superimposed on the stack, the stack being "processed" from below via the friction wheels 102, 103, so to speak.

When entering the space between the two friction wheels 102, 103, the tablet 13b has pressed them so far apart that the tablet 13b can pass through. Since the friction wheels 102, 103 are driven, they now take the tablet 13b with them and give it a certain speed, which moves it downward in a defined manner when it exits the release device 101.

The release device 101 therefore first brakes the tablets and then releases them at a defined acceleration / speed. The distance between the friction wheels 102, 103 and the measuring station 18 is just large enough that tablet dust which arises when rubbing does not impair the measurement. On the other hand, the distance is so small that the speed of the tablets 13 when passing through the measuring planes 37, 38 is exactly known. For this reason, the arrangement 57 from FIG. 5 or the spatially spaced arrangement of the measurement planes 37, 38 according to FIG. 4 can be dispensed with, although it can serve as an additional measure to carry out “final corrections”. The arrangement 57 and the release device 101 thus both serve, albeit in different ways, to specify or determine the falling speed of the objects 13 in the area of the measuring planes 37, 38. In the chute 14, a minimum backflow of tablets 13 may be required so that the release device 101 works properly. Heavy tablets that fall freely through the chute 14 may not be completely decelerated and passed on in a defined manner by the two friction wheels 102, 103. Thus, while there is a tablet between the friction wheels 102, 103, further tablets lie on this tablet 13b in the stack, which successively pass through the friction wheels 102, 103. If the back pressure falls below the minimum, which can be checked by a light barrier above the friction wheels 102, 103 in the chute 14, the friction wheels 102, 103 are stopped and the delivery is only continued when this minimum backflow is reached again.

The measuring device 18 shown comprises the lighting device 31 and the detector device 32 in the manner already described. However, the lighting device 31 is here a kind of "laser light curtain" which is generated by a laser source 111, the beam of which may be widened by means of optics, so that he spans a measuring plane, here measuring plane 37. In the case shown, the laser light source 111 is a clocked source, which clocks at approximately 50 kHz and works in the red or infrared wavelength range.

A sum sensor 112 is provided as the receiver, which, for example, has an approximately 25 mm wide reception slot and outputs a pure analog signal. A 5 kHz low-pass filter is available at the output of the receiver. A summation of individual signals per time cut is no longer necessary since the output is integrating over the entire receive slot, so to speak. If the receiving slot is completely covered, an output signal of 10 volts is obtained, for example, and if the receiving slot is completely free, an output signal of 0 volts is obtained. Because of the great coherence of the laser beam, the transmitter and receiver can be at a greater distance from one another than in the measuring stations 18 described so far. This distance can be up to 10 m, for example, which may be desirable in particular for geometric reasons. Furthermore, both measurement planes can now be arranged in one spatial plane without any problems, which considerably reduces the overall depth of the new device 10.

The interrogation of the sum sensor 112 can take place in such a way that depending on the size of the tablet 13 in the direction of fall 15 e.g. 50 measured values per object 12 can be taken. This resolution, which depends both on the interrogation frequency of the sum sensor 112 and on the speed of movement of the tablet 13 in the falling direction 15, is so good that tablets consisting of several segments can also be measured correctly. With this system it is not only possible to determine an integral area value (integration of support points), it is also possible to map the tablet by joining the individual "cuts", e.g. can be generated in a computer, as has already been explained with reference to FIG. 3, the sum sensor 112 then possibly having to be replaced by a line sensor.

In order to hide small dust particles etc., a lower threshold value is provided for the output signal of the sum sensor 112, which has to be exceeded several times in succession so that the object 12 passing through the measurement plane 37 is recognized and processed as a tablet 13 or as a fragment. In this way it is ensured that small dust particles are not ejected, but only fragments of a certain size.

Claims

1. Device for checking and / or distinguishing falling objects (12, 13), preferably free-falling tablets (13), with:

a lighting device (31), which is set up to generate a first intensity distribution (Jl) of light in a first measuring plane (37), which the respective object (12, 13) passes through the first measuring plane (37) influenced, and in a second measuring plane (38) different from the first measuring plane (37) to produce a second intensity distribution (J2) of light which influences the respective object (12, 13) when crossing the second measuring plane (38),

a detector device (32), which is set up to influence the first intensity distribution (J1) along a first measurement coordinate (48) and the influence of the second intensity distribution (J2) along one of the first measurement coordinate (48) in successive cuts to measure various second measurement coordinates (49) in such a way that it detects the respective dimensions of the object (12, 13) in the direction of the respective measurement coordinates (48, 49), and an evaluation device (19) which derives at least one control signal (S1, S2) from the detected dimensions, which contains a qualification of the respective object (12, 13), wherein

a switchable sorting station (21) forwards the objects (12, 13) differently depending on the control signal (S1, S2).

2. Device according to claim 1, characterized in that the first and the second measuring coordinate (48, 49) are approximately perpendicular to each other.

3. Apparatus according to claim 1 or claim 2, characterized in that the detector device (32) comprises at least one line sensor (41, 42; 66) with side by side arranged optical sensors.

4. Device according to one of claims 1 to 3, characterized in that the first and the second measuring plane (37, 38) run parallel to one another or preferably lie in the same spatial plane.

5. Device according to one of claims 1 to 4, characterized in that the detector device (32) comprises an imaging device (43, 44) which maps the intensity distribution (Jl, J2) on sensors.

6. The device according to claim 5, characterized in that the imaging device (43, 44) comprises at least one imaging light guide arrangement (64).

7. The device according to claim 5 or 6, characterized in that the lighting device (31) has at least one infrared light source (33), and that the imaging device comprises at least one infrared filter (45).

8. Device according to one of claims 1 to 7, characterized ge indicates that the lighting device (31) in the respective measuring plane (37, 38) lying, cell-shaped light transmitter (34, 36).

9. Device according to one of claims 1 to 8, characterized in that it comprises a device (57, 101) which is set up to specify or determine the speed at which the objects (12, 13) move in the area of the measuring planes (37, 38) move.

10. Device according to one of claims 1 to 9, characterized in that it is assigned to a counting station (93) which uses the control signal (Sl, S2) as a counting pulse.

11. Device according to one of claims 1 to 10, characterized in that the evaluation device (19) carries out at least one comparison with at least one reference signal (R) and generates the control signal (S1, S2) as a function of the comparison.

12. The device according to one of claims 1 to 11, characterized in that the objects (12, 13) maintain their orientation transverse to their direction of movement (15) at least when crossing the measurement planes (37, 38). 13. Device according to one of claims 9 to 12, characterized in that the device (57, 101) further comprises a release device (101) for the defined and individual release of the objects (12, 13) in the direction of fall

(15) the objects (12,

13) upstream of the measurement planes

(37, 38) is arranged.

14. The apparatus according to claim 13, characterized in that the release device (101) comprises at least one friction wheel (102, 103) which is resiliently mounted transversely to the direction of fall (15).

15. The apparatus according to claim 14, characterized in that the friction wheel (102, 103) is driven.

16. The device according to one of claims 13 to 15, characterized in that two friction wheels (102, 103) are provided.

17. Device according to one of claims 9 to 16, characterized in that the device (57, 101) comprises an arrangement (57) which determines or measures the speed at which the objects (12, 13) in the region of the measuring planes (37, 38) move.