Optical sensor for detecting a characteristic of a rotating blade, cutting unit comprising the optical sensor and method for detecting a characteristic of a rotating blade using the optical sensor

Abstract

An optical sensor (6, 7) to detect a characteristic of a rotating blade (4) comprises a U-shaped support (8) defining opposed ends between which the rotating blade can be inserted, an emitter (11) and a receiver (12) arranged in the U-shaped support to emit and receive a light radiation to be transmitted between the opposed ends of the U-shaped support. The receiver comprises an array of photodiodes (17) aligned in a radial direction with respect to the rotating blade. A processing unit (15) is connected to the receiver to receive a signal indicating the intensity of the light radiation received and to determine the characteristic of said rotating blade. A cutting unit (1) to cut a plate (2) of semiconductor material includes the rotating blade and such an optical sensor and can include an additional optical sensor to determine a position of the rotating blade. A method for operating the optical sensor can include the step of creating a digital image of the edge development of the rotating blade and analysing such digital image to determine features of the rotating blade.

Description

[0001] DESCRIPTION

[0002] "OPTICAL SENSOR FOR DETECTING A CHARACTERISTIC OF A ROTATING BLADE, CUTTING UNIT COMPRISING THE OPTICAL SENSOR AND METHOD FOR DETECTING A CHARACTERISTIC OF A ROTATING BLADE USING THE OPTICAL SENSOR"

[0003] TECHNICAL FIELD

[0004] The present invention relates to an optical sensor for detecting a characteristic of a rotating blade, a cutting unit comprising the optical sensor and a method for detecting a characteristic of a rotating blade using the optical sensor.

[0005] The present invention can be advantageously used in cutting units for cutting plates of semiconductor material or other materials used as substrates for the manufacture of electronic components.

[0006] The description below makes explicit reference to this example application, while remaining generally applicable.

[0007] BACKGROUND ART

[0008] A cutting unit used to cut a plate of semiconductor material (for example silicon) or metal, ceramic or glass materials used as substrates for making electronic components is provided with a rotating blade with a thin, continuous, non-serrated profile, and a platform to which the plate to be cut is fastened. The platform is moved in two dimensions to bring the rotating blade into contact with the plate, cutting it along suitable paths to separate circuits pre-printed thereon.

[0009] The cutting unit can be provided with two types of optical sensors: the first optical sensor is commonly referred to as the blade set-up sensor and measures the positioning of the rotating blade, in particular the distance of the edge of the blade from a predetermined reference position integral with the platform, and the second optical sensor is commonly referred to as the blade break sensor and detects any breakage of the rotating blade by detecting holes and / or irregularities in the edge of the rotating blade.

[0010] In general, both the optical blade set-up sensor and the optical blade break sensor comprise a receiver unit and an emitter unit: the emitter unit emits a beam of light radiation which is detected by the receiver unit based on a single photosensitive element. The rotating blade intercepts the light radiation emitted by the emitter unit and partially obscures the light radiation directed towards the receiver unit. Processing algorithms can be used to determine the position of the rotating blade from the obscuring of the light radiation, and irregularities and missing parts (holes) in the rotating blade can also be detected.

[0011] It is important to note that, to date, both in the case of an optical blade set-up sensor and in the case of an optical blade break sensor, the output from the receiver unit to a control unit of the cutting unit is a continuous analogue signal.

[0012] A cycle with the optical blade set-up sensor is used to calibrate the rotating blade after a new blade has been installed or to check the position of the rotating blade once a certain period of time has elapsed since the cutting of the plate of semiconductor material was started. Before the measurement, the optical blade set-up sensor is washed by spraying water and drying with a jet of air, after which the optical blade set-up sensor remains stationary and the rotating blade is moved by a spindle to a measurement position.

[0013] Generally, each cutting unit is provided with both an optical blade break sensor and an optical blade set-up sensor, and the cycle of the optical blade break sensor is always active during cutting, and the cutting unit is stopped if irregularities are indicated in the rotating blade.

[0014] Typically, during the whole cutting process an aqueous coolant is continuously sprayed on the rotating blade by means of properly oriented nozzles. The presence of the coolant can hinder the proper detection by the optical blade break sensor, since it is generally hard to distinguish breakages / irregularities of the rotating blade from water splashes.

[0015] The breakage of the rotating blade may also cause the optical blade break sensor to break, and therefore the optical blade break sensor must be easy to replace.

[0016] Finally, there is limited space available on the cutting unit to install optical sensors, and said sensors must therefore be small.

[0017] DISCLOSURE OF THE INVENTION

[0018] The object of the present invention is to provide an optical sensor for detecting a characteristic of a rotating blade that has a particularly compact layout, allows to perform very precise measurements even in a working environment involving, for instance, the presence of coolant and at the same time is easy and economical to produce.

[0019] Another object of the present invention is to provide a cutting unit comprising an optical sensor for detecting a characteristic of a rotating blade.

[0020] A further object of the present invention is to provide a method for detecting a characteristic of a rotating blade by using an optical sensor.

[0021] According to the present invention, an optical sensor for detecting a characteristic of a rotating blade, a cutting unit comprising the optical sensor and a method for detecting a characteristic of a rotating blade by using an optical sensor according to the attached claims are provided.

[0022] The claims describe preferred embodiments of the present invention and are an integral part of the present description.

[0023] BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The present invention is described below with reference to the attached drawings, which show some non-limiting example embodiments of the invention, in which:

[0025] • Figure 1 is a schematic view of a cutting unit provided with two optical sensors according to the present invention;

[0026] • Figure 2 is a schematic view of an optical sensor from Figure 1;

[0027] • Figure 3 is a schematic view of a different embodiment of an optical sensor from Figure 1;

[0028] • Figure 4 is a graph showing the measurements of an array of photodiodes of an optical sensor from Figure 1;

[0029] • Figures 5, 6 and 7 are schematic representations of the readings from an array of photodiodes of an optical sensor from Figure 1;

[0030] • Figures 8, 9 and 10 are graphs showing the evolution of the measurements of an optical sensor from Figure 1, with no problems, with a partial breakage of a rotating blade, and with an eccentricity in the rotating blade, respectively; and

[0031] • Figures 11 and 12 are graphs showing the evolution of the measurements of an optical sensor, corresponding to Figures 9 and 10, respectively, when environment noise is present.

[0032] BEST MODES FOR CARRYING OUT THE INVENTION

[0033] In Figure 1, reference number 1 denotes a cutting unit as a whole for cutting a plate 2 of semiconductor material used as a substrate for the manufacture of electronic components. According to other embodiments, the plate 2 could be made of a type of material other than semiconductor material, and could therefore for example be a plate of metal, ceramic or glass material, preferably used as a substrate for the manufacture of electronic components.

[0034] The cutting unit 1 comprises a platform 3 on which the plate 2 of semiconductor material is mounted, a (circular) rotating blade 4, and a spindle 5 with a rotating shaft configured to support the rotating blade 4. The spindle 5 is movable so as to displace the rotating blade 4 to a working position (illustrated with an unbroken line in Figure 1) in which the rotating blade 4 is arranged to interfere with the plate 2 of semiconductor material to cut, during mutual movements between the spindle 5 and the platform 3, the plate 2 of semiconductor material, and to a calibration position (illustrated with a dashed line in Figure 1) in which the rotating blade 4 is relatively far from the plate 2 of semiconductor material to check the position of the rotating blade 4 or calibrate (i.e. reset the position of) the rotating blade 4 with respect to a reference position integral with the platform 3.

[0035] The cutting unit 1 comprises an optical sensor 6, or first optical sensor 6, which is paired with the rotating blade 4 at least when the rotating blade 4 is in the working position (illustrated with an unbroken line in Figure 1) and is used to measure wear (breakage, chipping, etc.) of the rotating blade 4 (i.e. to detect irregularities and missing parts in the edge of the rotating blade 4). The optical sensor 6 can be integral with the spindle 5 (i.e. with the support of the rotating blade 4 and therefore be movable together with the spindle 5 and with the rotating blade 4) as illustrated in Figure 1.

[0036] The cutting unit 1 comprises an optical sensor 7, or second optical sensor 7, that is paired with the rotating blade 4 only when the rotating blade 4 is in the calibration position (illustrated with a dashed line in Figure 1) and is used to calibrate the rotating blade 4 so measuring or determining the position of the rotating blade 4 with respect to a preset reference position of the platform 3 in a known position with respect to the plate 2 being cut. In particular, the optical sensor 7 is integral with the platform 3 such that the position determined by the optical sensor 7 relates directly to the platform 3 (unlike the position determined by the optical sensor 6 which relates to the spindle 5).

[0037] The optical sensor 7 is used to calibrate the rotating blade 4 after a new blade has been installed or to check the position of the rotating blade 4 once a certain period of time has elapsed since the cutting of the plate 2 of semiconductor material was started. Before the measurement, the optical sensor 7 is washed by spraying water and dried with a jet of air, after which the optical sensor 7 remains stationary and the rotating blade 4 is moved by the spindle 5 to the calibration position (illustrated with a dashed line in Figure 1).

[0038] Typically, the measurement cycle for detecting the wear of the rotating blade 4 is always active during cutting, and the cutting is stopped if wear of the rotating blade 4 is indicated.

[0039] The optical sensors 6 and 7 are structurally identical (and therefore potentially interchangeable) and the measurements provided thereby are processed differently to obtain different information (as explained above).

[0040] Figures 2 and 3 schematically illustrate two different embodiments of the optical sensor 6 or 7 (identical parts in the two embodiments are indicated using identical reference numbers).

[0041] The optical sensor 6 or 7 comprises a U-shaped support 8 (i.e. fork-shaped with two separate prongs 9 and 10) that is preferably made of stainless steel. The support 8 comprises the two opposing prongs 9 and 10, between which, i.e. between the opposed ends of the U-shaped support 8 (or "opposed ends of the U-shape"), the rotating blade 4 can be inserted when in use. In other words, when in use, the U-shaped support 8 and the spindle 5 are arranged so that the rotating blade 4 is between the two opposing prongs 9, 10, between the opposed ends of the U-shape.

[0042] The optical sensor 6 or 7 comprises an emitter 11 and a receiver 12, arranged in the U-shaped support 8 and configured to emit and receive a light radiation that is transmitted between the opposed ends of the U-shape. In particular, in the embodiment in Figure 2, the emitter 11 (for example provided with an LED or a laser) is arranged on the prong 9 and is configured to emit light radiation directed towards the prong 10, and the receiver 12 is arranged on the prong 10 and is configured to receive the light radiation and to measure an intensity of the received light radiation, that is to provide a signal indicative of the intensity of the received light radiation. In Figures 2 and 3 the receiver 12 is shown very schematically, highlighting particularly important features of the present invention. However, components such as lenses / optical assemblies, which may generally be provided to improve the optical quality of the shadow received, are not represented. The emitter 11 comprises a dedicated control circuit 13 which controls the emission of the light radiation. The receiver 12 comprises a dedicated control circuit 14 which processes the measurement of the received light radiation, preferably by converting the analogue signal into a digital signal. For example, the lightradiation intensity measurement is converted by a compact microcontroller using its integrated analogue-to-digital converters (ADCs), although external converters outside the microcontroller can also be used. According to a preferred embodiment illustrated in the attached figures, the control circuit 13 of the emitter 11 and the control circuit 14 of the receiver 12 are housed inside the support 8 of the optical sensor 6 or 7.

[0043] As illustrated in Figure 1, each optical sensor 6 or 7 has a dedicated processing unit 15 (i.e. a control board) which is arranged outside the support 8 and is electrically connected to the emitter 11 and the receiver 12, more specifically to the control circuits 13 and 14 of the emitter 11 and of the receiver 12, by an electrical cable 16 (i.e. a cable carrying an electrical signal, which is preferably digital but may also be analogue). The processing unit 15 of each optical sensor 6 or 7 processes the intensity measurement of the light radiation received by the receiver 12 to extract the desired information, that is to detect the characteristic of the rotating blade. The processing unit 15 of the optical sensor 6 is primarily configured to detect wear (breakage, chipping, etc.) of the rotating blade 4 while the processing unit 15 of the optical sensor 7 is configured to measure the position of the rotating blade 4 with respect to a predefined reference position of the platform 3.

[0044] According to an alternative embodiment, there is a single processing unit 15 which is connected to both optical sensors 6 and 7, is common to both optical sensors 6 and 7, and processes the measurements from the two optical sensors 6 and 7 alternately or simultaneously.

[0045] According to an additional embodiment, the processing unit 15 is arranged inside the support 8 of the optical sensor 6 or 7 and therefore all the components of the optical sensor 6 or 7 are contained in the support 8.

[0046] As illustrated in Figures 2 and 3, the receiver 12 comprises an array (row) of light radiation sensitive elements, e.g. photodiodes 17, typically more than sixteen photodiodes, for example thirty-two photodiodes, that are aligned in a radial direction with respect to the rotating blade 4 (i.e. a direction perpendicular to an axis of rotation of the rotating blade 4). In other words, the array of photodiodes 17 has a single row of photodiodes 17 arranged on a given line oriented radially with respect to the rotating blade 4. Each photodiode 17 is configured to measure the intensity of the light radiation received by the photodiode 17, independently of the other photodiodes 17, i.e. to provide, on a scale (for example between 0 and 100), a measurement of the intensity of the light radiation received. The receiver 12 therefore outputs a number of measurements of light intensity equal to the number of photodiodes 17 (for example thirty-two). According to other entirely equivalent embodiments, the number of photodiodes 17 that make up the array may be different (i.e. there may be fewer photodiodes 17 or more photodiodes 17) and the minimum number of photodiodes 17 that make up the array is two (although there are usually at least ten photodiodes 17).

[0047] In general, the receiver 12 comprises an array (row) of at least two photodiodes 17 and, according to other embodiments (not illustrated), comprises a matrix of photodiodes 17 made up of at least two arrays (rows) of photodiodes 17 arranged side by side. In essence, the receiver 12 is a sensor formed by a line or grid (matrix) of semiconductor elements (photodiodes 17) that can accumulate an electrical charge proportional to the intensity of the electromagnetic radiation incident thereon. It may be a charge-coupled device (CCD) circuit or use a different technology, such as CMOS sensor technology. By sending a timed sequence of pulses to the receiver 12, an electrical signal is output and it is possible to reconstruct the matrix of the pixels that make up the projected image on the surface of the receiver 12.

[0048] The emitter 11 illustrated in Figure 2 has a small size and can be housed in the prong 9 of the support 8. In addition, preferably, the emitter 11 has high directivity, i.e. the scattering angle of the rays of the light radiation is reduced in order to cast a sharp shadow on the receiver 12.

[0049] In the embodiments illustrated in Figures 2 and 3, the emitter 11 comprises a light source 18 which is housed in the support 8 and preferably comprises an LED of very small mechanical dimensions.

[0050] In the embodiment illustrated in Figure 2, the light source 18 (for example an LED) of the emitter 11 directly emits the light radiation perpendicular to the prong 9 and directed towards the prong 10. The advantage of this embodiment is the simplicity of implementation but, on the other hand, it requires a light source 18 with a small emitting area and polar emission large enough to cover the entire field of the receiver 12, while also being of suitably small mechanical dimensions in the emission direction.

[0051] In the embodiment illustrated in Figure 3, the light source 18 of the emitter 11 emits the light radiation parallel to the prong 9, i.e. along a direction substantially radial with respect to the rotating blade 4, towards one of the two opposed ends of the U-shaped support 8, where there is an optical deflecting element 19 (for example a prism with a reflective surface 19', or a mirror) that reflects the light radiation emitted by the light source 18 perpendicular to the prong 9 and directed towards the prong 10, i.e. towards the other of the two opposed ends of the U-shaped support 8. Figure 3 shows the components very schematically to illustrate as clearly as possible (and in the space available) important aspects of the present invention, disregarding functional coherence in some cases. This applies in particular to the optical deflecting element 19 which is shown in Figure 3 with a reflective surface 19' inclined with respect to the prong 9 and the emission direction of the light source 18 by an angle that is clearly not 45°, whereas this angle must be (and is) 45° to properly reflect (as shown graphically in Figure 3) the light radiation perpendicular to the prong 9. This embodiment, while structurally more complex, has the advantage of better defining the shadow cast because it enables sources with smaller emission areas and narrow polar emission (i.e. highly directed) to be used, since there is enough space (distance) for the beam to widen enough to cover the entire array of photodiodes 17. In this embodiment, mechanically larger sources can be used, thus providing a wider selection of available light sources 18 (for example LEDs) with high directivity. In addition, this embodiment enables a lens to be inserted in the optical path to collimate the light radiation.

[0052] As illustrated in Figure 1, the two processing units 15 of the sensors 6 and 7 are connected to a control unit 20 of the cutting unit 1, to which they provide the results of the measurements.

[0053] Figure 4 is a graph illustrating the measurements of an array of thirty-two photodiodes 17 of a receiver 12. As shown, the first fourteen photodiodes 17 are substantially in shadow (i.e. the light from the emitter 11 directed towards the first fourteen photodiodes 17 is blocked by the rotating blade 4), the last fifteen photodiodes are in full light (i.e. the light from the emitter 11 directed towards the last fifteen photodiodes 17 is not blocked at all by the rotating blade 4) and the three photodiodes 17 in the middle are partially illuminated (i.e. the light from the emitter 11 directed towards the three photodiodes 17 in the middle is partially blocked by the rotating blade 4).

[0054] According to a possible embodiment illustrated schematically in Figures 5, 6 and 7, the processing unit 15 is configured to detect the desired characteristic of the rotating blade 4 by considering only the light-radiation intensity measurements provided by a group G made up of a limited number of the photodiodes 17 (in the most limited case, by a single photodiode 17), that are located around an edge of the rotating blade 4 (i.e. straddling the shadow zone in which the rotating blade 4 blocks the passage of light radiation and the light zone in which the rotating blade 4 does not block the passage of light radiation). In this embodiment, the processing unit 15 sums the light-radiation intensity measurements provided by the group G made up of a limited number of photodiodes 17 (or by the single photodiode 17) that are located around an edge of the rotating blade 4 (i.e. straddling the shadow zone and the light zone), compares this measurement with a predetermined intensity measurement, preferably with the total intensity measurement of the group G of photodiodes 17 (or of the single photodiode 17) previously taken when the photodiodes 17 in the group G (or the single photodiode 17) were all illuminated and, on the basis of a predetermined threshold, identifies the position of the rotating blade 4.

[0055] In particular, given that the position of the photodiodes 17 is known, the position of the edge (corresponding to the external radius) of the rotating blade 4 is determined by identifying the last photodiode 17 that is not obscured (or the first photodiode 17 that is obscured) on the basis of a predetermined threshold. Then the light-radiation intensity measurements provided by the photodiodes 17 adjacent to said last / first photodiode 17 that are located around the edge of the rotating blade 4 being measured are added together and the percentage of intensity of the obscured light radiation, and therefore the measurement of a position of the edge of the rotating blade 4, is obtained. The embodiment described above with reference to figures 5 to 7 is particularly useful in situations where it is not possible to move the optical sensor 6 or 7 in relation to the rotating blade 4, and provides a measurement with a higher resolution than the measurement that would be obtained by adding together the measurements of all the photodiodes 17 of the array, as well as a measuring range as large as the size of the array, since the length of the measurement area can be selected by simply adding the intensity measured by the limited number of photodiodes 17 (or, at the extreme, the single photodiode 17) of the group G and ignoring photodiodes that do not provide any information for the measurement because they are very far away from the dark / light transition zone.

[0056] Figures 5, 6, and 7 show three examples where the group G is made up of a limited number of photodiodes 17 that straddle the edge of the rotating blade 4 (i.e. that straddle the shadow zone and the light zone). In the example embodiment illustrated in Figures 5, 6 and 7, the limited number of photodiodes 17 that make up the group G is three, but this number may be different (typically between one and five).

[0057] According to an alternative embodiment, the optical sensor 6 or 7 comprises a first array of photodiodes 17 and a second array of photodiodes 17. The last obscured photodiode 17 is identified in the first array of photodiodes 17, and the first obscured photodiode 17 is identified in the second array of photodiodes 17, or vice-versa, thereby providing two different types of detection that can be averaged to increase the precision and confidence level of the measurement.

[0058] According to a possible embodiment, due to the high resolution for receiving the light radiation, the optical sensor 6 can also be used to detect a variation of the position of the edge of the rotating blade, more specifically to take measurements of the position of the rotating blade 4 in relation to a predefined reference position of the spindle 5 (the position of which is measurable with respect to the platform 3) and therefore can replace, partially or completely, the function of the optical sensor 7, or at least cooperate with the optical sensor 7, for instance as follows: when in use (i.e. during cutting), the optical sensor 6 takes measurements of the position of the rotating blade 4 and these measurements are used to determine if and when the position of the rotating blade 4 with respect to the platform 3 needs to be recalibrated using the optical sensor 7. In other words, taking measurements of the position of the rotating blade 4 using the optical sensor 6 makes it possible to predict (determine) if and when the position of the rotating blade 4 with respect to the platform 3 needs to be recalibrated using the optical sensor 7. As a consequence, the calibration of the position of the rotating blade 4 with respect to the platform 3 by the optical sensor 7 must not necessarily be performed at regular intervals (which is usually too often, thereby reducing the productivity of the cutting unit 1) but can only be performed when it is actually needed. That is, the processing unit 15 of the optical sensor 6, based on the light intensity measurement processed, is able to provide a signal relating to the need / opportunity to carry out a new calibration of the rotating blade 4 using the optical sensor 7. In other words, the detections performed by the optical sensor 6 can be used to detect a variation in the position of the edge of the rotating blade 4 and this variation (when it exceeds a certain threshold) indicates the need for / timeliness of a recalibration of the rotating blade 4 using the optical sensor 7. In the above-mentioned embodiment where a single processing unit 15 is connected to both optical sensors 6 and 7, the measurements of the two optical sensors 6 and 7 can be simultaneously processed so as to advantageously obtain an instantaneous and synergic combination of the two position measurements.

[0059] The processing unit 15 of the optical sensor 6 or 7 is configured to create a digital image (illustrated by way of example in Figures 8, 9 and 10) of the development of the edge of the rotating blade 4 using estimates of the position of the edge of the rotating blade 4 for at least a full rotation of the rotating blade 4 (i.e. for a 360° rotation of the rotating blade 4). Then, the processing unit 15 of the optical sensor 6 or 7 analyses (uses) the digital image of the development of the edge of the rotating blade 4 to detect at least one characteristic of the rotating blade 4, such as eccentricity, run-out, or the like, in a known manner.

[0060] In this regard, it should be noted that the angular position of the rotating blade 4 can be detected by an angle encoder paired with the rotating blade 4 and therefore be known with high precision and good resolution. In other words, the digital image of the development of the edge of the rotating blade 4 is obtained from the estimates of the position of the edge of the rotating blade 4 over a period of time in which the rotating blade 4 makes at least a full rotation. Where high precision is not required, it is still possible, even without an encoder, to obtain the digital image of the development of the edge of the rotating blade 4 by indirect correlation, for example by identifying a period of rotation of the rotating blade 4 from an analysis of the trend of the acquired signal over time.

[0061] Figure 8 illustrates an example of a digital image of the development of the edge of a problem free rotating blade 4. As shown, the position of the edge of the rotating blade 4 is (substantially) constant throughout the rotation of the rotating blade 4.

[0062] Figure 9 illustrates an example of a digital image of the development of the edge of the rotating blade 4 that has a break (i.e. a missing part). As shown, the position of the edge of the rotating blade 4 has a "hole" indicating a break (i.e. a missing part). The angular width of the "hole" is proportional to the circumferential dimension of the break, while the depth of the "hole" is proportional to the radial dimension of the break.

[0063] Figure 10 illustrates an example of a digital image of the development of the edge of the rotating blade 4 with an eccentricity (for example due to uneven wear). As shown, the position of the edge of the rotating blade 4 is not constant but varies in an approximately sinusoidal pattern. That is, in the case of eccentricity of the rotating blade 4, the position of the edge of the rotating blade 4 appears as a curve and not as a straight line.

[0064] In other words, the array of photodiodes 17 of the receiver 12 enables the position of the edge of the blade 4 to be determined punctually and the position of the edge of the blade 4 in space and time can be processed to obtain a two-dimensional digital image (illustrated by way of example in Figures 8, 9 and 10) of the edge of the rotating blade 4.

[0065] The digital image has high spatial and temporal resolution, and can be advantageously used to carry out advance filtering processes allowing to get reliable information from the acquired measurement signals even where the working environment involves hard operating conditions. For example, the aqueous coolant that is continuously sprayed on the rotating blade 4 and consequent splashes generates a numeric noise that interferes with the useful signal detected by the sensor 6, so jeopardizing the proper detection of any breakages / irregularities of the rotating blade 4. Figures 11 and 12 correspond, respectively, to Figures 9 and 10 and show the acquired digital image of the measurement signal where numeric noise generated by the coolant splashes is present. The processing unit 15 can store and manage a high number of measurement signals provided by the array(s) of photodiodes 17 in a time interval corresponding to at least one full rotation of the rotating blade 4. Contrary to what happens in known sensors having a single photodiode, the presence of a number of photodiodes 17 guarantees a high spatial resolution and allows to identify different spatial evolution lines. For instance, it is consequently possible to identify the trend of a signal corresponding to the numeric noise generated by the coolant splashes and to distinguish it from the measurement signal corresponding to breakages / irregularities of the rotating blade 4, since the two events have each a specific spatial and temporal fingerprint that can be identified by the processing unit 15. Moreover, when the acquisition time interval allows the rotating blade 4 to perform more full rotations, detecting the periodic occurrence of events that are not correlated with the coolant splashes allows to tune the filtering with the rotation speed of the rotating blade 4 so improving the identification of the proper measurement signal. In other words, the analysis of the digital image can include the application of filtering processes allowing, for example, to identify the proper measurement signal relating to the characteristics of the rotating blade and distinguish it from numeric noise.

[0066] The optical sensor 6 or 7 described above has numerous advantages.

[0067] In particular, the optical sensor 6 or 7 described above has a higher resolution in detecting the received light radiation due to the fact that the receiver 12 comprises an array of photodiodes 17 (i.e. a plurality of photodiodes 17 arranged in a straight line). Among other things, using more photodiodes 17 (sensitive elements) enables the application of new, more flexible, more advanced processing algorithms, both for detecting alert thresholds and for taking actual dimensional measurements. An additional advantage of the optical sensor 6 or 7 described above is its extreme miniaturization, which is also achieved by integrating the control circuits 13 and 14 into the support 8.

[0068] Finally, the sensor 6 or 7 described above is simple and inexpensive because it involves the use of components that are easily available on the market at low cost.

Claims

CLAIMS1. Optical sensor (6, 7) to detect a characteristic of a rotating blade (4) and comprising : a U-shaped support (8) defining two opposed ends between which the rotating blade (4) can be inserted; an emitter (11) and a receiver (12) arranged in the U-shaped support (8) and configured to emit and to receive a light radiation that is transmitted between the opposed ends of the U-shaped support; and a processing unit (15) connected to the receiver (12) to receive a signal indicative of the intensity of the received light radiation and detect the characteristic of the rotating blade (4); the optical sensor (6, 7) being characterized in that the receiver (12) includes at least an array of light radiation sensitive elements (17) which are aligned in a radial direction with respect to the rotating blade (4).

2. Sensor (6, 7) according to claim 1, wherein each sensitive element (17) is configured to measure an intensity of the light radiation received by the sensitive element (17) independently of the other sensitive elements (17).

3. Sensor (6, 7) according to claim 1 or claim 2, wherein each sensitive element (17) is a photodiode.

4. Sensor (6, 7) according to any one of claims 1 to 3, including a control circuit (13) of the emitter (11) and a control circuit (14) of the receiver (12) housed in the support (8) and electrically connected to the processing unit (15).

5. Sensor (6, 7) according to any one of claims 1 to 4, wherein the emitter (11) includes a light source (18) that is arranged in the U-shaped support (8).

6. Sensor (6, 7) according to claim 5, wherein the light source (18) emits the light radiation along a direction substantially radial with respect to said rotating blade (4) towards one of the two opposed ends of the U-shaped support (8) the emitter (11) including an optical deflecting element (19) which reflects the light radiation emitted by the light source (18) towards the other of the two opposed ends of the U-shaped support (8).

7. Sensor (6, 7) according to any one of claims 1 to 6, wherein the processing unit (15) is configured to determine the feature of the rotating blade (4) by taking into account only intensity values of the light radiation provided by a group (G) made up of a limited number of sensitive elements (17) that are located between a shadow zone in which the rotating blade (4) blocks the passage of a light radiation and a light zone in which the rotating blade (4) does not block the passage of the light radiation.

8. Sensor (6, 7) according to claim 7, wherein the receiver (12) includes more than sixteen sensitive elements (17) and said group (G) is made up of a number of sensitive elements (17) comprised between one and five.

9. Sensor (6, 7) according to claim 7 or claim 8, including a first array of sensitive elements (17) in which the last obscured sensitive element (17) is identified and a second array of sensitive elements (17) in which the first obscured sensitive element (17) is identified.

10. Sensor (6, 7) according to any one of claims 1 to 9, wherein the receiver (12) includes a matrix of light radiation sensitive elements (17) made up of at least two side by side arrays of light radiation sensitive elements (17).

11. Cutting unit (1) to cut a plate (2) of semiconductor material comprising:- a rotating blade (4)- a spindle (5) configured for supporting the rotating blade (4); and- a first optical sensor (6, 7) according to any one of claims 1 to 10 that can be coupled to the rotating blade (4) and can be used to detect a characteristic of the rotating blade (4).

12. Cutting unit (1) according to claim 11, wherein the first optical sensor (6) is also used to detect a variation of the position of the edge of the rotating blade (4).

13. Cutting unit (1) according to claim 11 or claim 12 and including a second optical sensor (7) according to any one of claims 1 to 10 that can be used to calibrate the rotating blade (4) so determining a position of the rotating blade (4) with respect to a preset reference position.

14. Cutting unit (1) according to claim 13 as depending on claim 12, wherein the first optical sensor (6) is used to report the need / opportunity to carry out a new calibration of the rotating blade (4) by means of the second optical sensor (7).

15. Cutting unit (1) according to claim 13 or claim 14, wherein the spindle (5) is movable so as to displace the rotating blade (4) to a working position in which the rotating blade (4) is paired only with the first optical sensor (6) and in a calibration position in which the rotating blade (4) is paired also with the second optical sensor (7).

16. Cutting unit (1) according to any one of claims 11 to 15, wherein the first optical sensor (6) is integral with the spindle (5).

17. Method for detecting a characteristic of a rotating blade (4) by means of the optical sensor (6, 7) according to any one of claims 1 to10 and including the following steps: arranging the U-shaped support (8) so that the rotating blade (4) is placed between the two opposed ends of the U-shaped support (8); emitting from the emitter (11) a light radiation directed towards the receiver (12); measuring, by means of the receiver (12), an intensity of the light radiation emitted by the emitter (11); and detecting the characteristic of the rotating blade (4) based on the measure carried out by the receiver (12).

18. Method according to claim 17 and including the steps of:- summing the measures of intensity of the light radiation provided by a group (G) made up of a limited number of sensitive elements (17) that are located between a shadow zone in which the rotating blade (4) blocks the passage of a light radiation and a light zone in which the rotating blade (4) does not block the passage of the light radiation;- comparing the sum of the measures of intensity of the light radiation provided by said group (G) of sensitive elements (17) with a predetermined intensity measurement; and- identifying, on the basis of a predetermined threshold, the position of the rotating blade (4).

19. Method according to claim 17 or claim 18 and including the steps of:- identifying the last sensitive element (17) that is not obscured or the first sensitive element (17) that is obscured on the basis of a predetermined threshold;- adding together the measures of intensity of the light radiation provided by the sensitive elements (17) adjacent to said last / first sensitive element (17) to obtain a percentage of obscured lightradiation; and- determining the position of the edge of the rotating blade (4) as a function of the percentage of obscured light radiation.

20. Method according to any one of claims 17 to 19 and including the steps of:- creating a digital image of the edge development of the rotating blade (4) by considering estimates of the position of the edge of the rotating blade (4) for at least a whole rotation of the rotating blade; and- analyzing the digital image of the edge development of the rotating blade (4) in order to detect characteristics of the rotating blade (4).

21. Method according to claim 20, wherein the step of analysing the digital image includes applying filtering processes allowing to identify the proper measurement signal relating to the characteristics of the rotating blade and to distinguish it from numeric noise.

22. Method according to claim 21, wherein the step of analysing the digital image includes applying filtering processes allowing to distinguish the proper measurement signal to numeric noise due to the presence of splashes of coolant that is sprayed on the rotating blade

Images