NL8600790A - METHOD AND APPARATUS FOR DETERMINING THE POSITION OF ERRORS IN FLAT GLASS. - Google Patents

Description

V

NL 33415-dV / i 'X 4

Method and device for determining the location of errors in flat glass.

The invention relates to a method for determining the location of errors in flat glass during a trajectory, in which the glass is scanned with a beam of electromagnetic radiation and a photodetector means using which the incident on the beam errors in the glass are detected. The invention also relates to a device for applying this method.

The invention particularly relates to determining the location of errors in a moving belt of glass, but may also be used to determine the location of errors in pre-aligned glass plates during their transportation along a predetermined path.

Today, flat glass is manufactured almost exclusively as a belt in a drawing machine or a float chamber, the belt being transported through a slow cooling furnace to a cutting station, where it is cut into plates for storage or further processing. This cutting sometimes takes place automatically under computer control, so that plates of the desired dimensions can be cut from the belt with a minimum glass loss.

The quality indication of glass plates depends on the number and severity of the errors in the plates. When a strip or sheet of flat glass is to be cut so that sheets of the desired dimensions and quality designation are obtained, it is desirable to predetermine the presence and location of significant errors so that their positions can be taken into account when determining the positions , where to cut the glass.

In the case of an industrial factory, it is highly desirable to perform the error detection procedure automatically, and automatic detection methods using the deflecting influence of errors on a scanning beam of light are already known. Reference is made to British Patent Specification 1,526,930, which describes a system in which a plate or 1. . - '*. * - 2 -, · ”Ν glass belt, moving over a conveyor, is repeatedly transversely scanned by a light beam and where deflections of the light beam due to the presence of errors are monitored by a photo-detector system, which detects suitable signals to a memory. The location of light-deflecting errors can be deduced from the time relationship of these signals. The number of the transverse scan or the scan time indicates the location of the longitudinal error of the glass and the transverse position is derived by comparing with a fixed position.

The known detection methods, which use a scanning light beam, can only determine the locations of errors with sufficient accuracy, if it is ensured that a side edge of the examined glass follows a constant line of movement during the entire scanning operation of the glass, otherwise, what is signaled is not the location of a transverse error of the glass, but the location relative to the width of the conveyor. Under conditions, the positioning of the workpiece cannot be adequately ensured. In some installations, a variation in the line of motion of the side edges of the moving glass is inevitable. An important example of this is a production installation for float glass. The position of the float glass relative to the width of the conveyor through which the glass is conveyed from the float reservoir and then through the cooling oven slowly cooling the glass is subject to variation over time due to changes in forces, which are applied to the glass belt when changing from one float glass thickness to another and due to "snaking" of the belt on the conveyor. The known detection methods therefore do not suffice in automatically detecting the location of errors in a properly formed band of float glass and there is a need for a method suitable for this purpose.

The object of the invention is to provide such a method.

\> /; <; M i ..- 3- «*

According to the invention, the method of the type mentioned in the preamble has the feature for this purpose that - the glass is scanned by a beam of electromagnetic radiation moving across the path of the glass, so that the beam follows successive transverse tracks across the glass and at each scan the beam passes through the opposite side edges of the glass; the photodetector means for detecting deflections of the beam caused by hitting one or each side edge of the glass as well as detecting attenuation or deflections of the beam due to errors in the glass? - which photodetector member forms part of a signaling device which supplies output signals, which indicate positions along the scanned length of the glass according to the transverse tracks, in which errors are detected and indicate the distances measured along these tracks between these errors and a side edge of the glass, and - wherein the output signals are used to identify the signaled error sites.

In this way, the important advantage is achieved that changes in the position of a side edge of the glass in the scanning station during the movement of the glass through that station are turned off in the sense that they do not affect the accuracy of the signaled error spots on the glass. The method according to the invention is therefore particularly suitable for use for automatically registering the locations of errors in a moving belt of newly manufactured flat glass. However, the invention can also be used to determine the location of errors in moving pre-oriented flat glass plates, the lateral positions of which on a conveyor are not accurately controlled.

By a suitable arrangement of the photodetection device, as described in more detail below, the method according to the invention can determine the location of errors of one or more different types. This includes errors, which I - 4 - s ,.

cause a deflection of an incident beam of electromagnetic radiation due to refraction. Such errors include bubbles, grains and stones, which mainly determine the quality indication of flat glass plates. The method can also detect impermeable zone errors. If, for example, use is made of a photoelectric detector arranged such that it is normally irradiated by the beam, but emits an error signal, if irradiation no longer takes place, because the beam 10 becomes more than a predetermined amount. deflected, the detector will necessarily provide a corresponding signal if the beam hits an impermeable zone in the glass. In addition or alternatively, the method can detect errors that cause significant attenuation of the scanning beam 15. Such errors include, for example, certain discolorations, such as tin discolorations from a metal bath.

For the sake of simplicity, positions along the scanned length of the glass of the transverse tracks in which errors are detected are hereinafter referred to as the longitudinal positions of the errors, while the transverse distances between the detected errors and a side edge of the glass are indicated as the cross ordinates of errors.

A particularly important practical use of the method according to the invention lies in the automatic control of a marker or glass cutter, which during the continuous movement of the glass past the detection station, where the glass is scanned, marks the glass at the location of the faults or the glass cuts at the places that concern the faults. According to a preferred embodiment of the invention, the signaling device generates outputs which are used as a control factor in automatically controlling a glass marking unit or cutter to mark or cut the moving glass at a location downstream of the station where the glass is placed. scanned. In such an embodiment of the invention, the longitudinal positions of errors can be easily predicted by the times at which error detection signals occur, since the time of arrival of an error in the marking or cutting station depends, on the one hand, on the time of the error detection and on the other hand the speed of movement of the glass and the distance between the marking or cutting station and the detection station where the glass is scanned.

As an alternative to the generation of output signals indicating the longitudinal positions of the errors by the times at which these signals occur, the signal means may generate output signals indicating the longitudinal distances between transverse tracks in which errors are generated and a transverse direction output line, This procedure can be used, for example, in detecting errors in moving glass plates by detecting the arrival of the leading transverse edge of each plate in the detection station and generating output signals, which are the longitudinal distances of the errors from this indicate the edge, which serves as the output line. Such output signals can be applied to a recording means which reproduces the information 20 in the form of a statistical recording for the glass quality classification or other purpose.

The transverse ordinate of an error can be determined by the signaling means directly from the scanning distance covered by the scanning beam between the time 25 (indicated by the occurrence of an edge deflection signal from the photodetector element) at which the beam arrives at the location of a side edge of the glass, and the time at that scanning pass, at which an error signal from the photodetector indicates that the beam is hitting the error. Alternatively, the transverse coordinate can be determined by the signaling device as a function of the time interval between the times of occurrence of the edge deflection and error signals, however, the scanning speed of the beam across the glass must be at least substantially constant.

It is not necessary that the signaling go always determine the cross-ordinate of an error based on a glass edge position signal or signals supplied at the actual beam passage or passages in which the error is hit. The glass length increment from scan to scan will be too small to cause a significant change in the position of the glass edges in the detection station 5 in time between two consecutive scan or even in time of a large number of consecutive scan. According to the invention, an average value can be derived from the glass edge deflection signals supplied by the photodetector at a number of consecutive passes of the beam across the glass in one direction, which average value is used in deriving the output signals. indicate the cross ordinates of errors.

Preferably, the error site output signals 15 are generated only for errors whose cross-ordinates (ordinates) are above a predetermined minimum value from each side edge of the glass. Edge parts of drawn or float glass often have an inferior quality and in determining where to cut this glass to obtain the required sheet sizes for sale, it is often necessary to treat the edge parts as waste. In the case of float glass, edge strips of the belt are spoiled by the action of top rollers in the float reservoir and these edge parts of the glass should be disregarded. By applying the aforementioned feature of the invention, the presence of errors in the edge parts of the glass is disregarded in the output signals, which serve as data for identifying the error sites. The width of the marginal parts to be disregarded is predetermined and this size is therefore available as a control parameter for the signaling member. The signaling device can be designed in such a way that such edge errors do not cause the generation of photoelectric detection signals or that such signals are generated, but do not result in output signals.

Preferably, the displacement of the beam is monitored by a monitor, which provides signals that monitor the

-. - "J

"V

indicate beam positions relative to a fixed starting position, which is outside the limits of the beam movement across the glass, which signals are used by the signaling device in deriving the output signals. The generation of output signals, which indicate the transverse positions of the side edges and of errors relative to a fixed starting position, is advantageous, because under these conditions the output signals are particularly useful for automatically controlling operations on the moving glass by in a direction mounted at a predetermined fixed position downstream of the location where the glass is scanned.

In the manufacture of a continuous belt of glass, the edge parts of which must be removed as waste, it is advantageous to cut the edges of the belt by means of cutting members, which are controlled depending on the variations in the positions side edges of the tire, so that such variations do not result in significant variations in the width of the removed edges. In order to achieve this advantage, in an embodiment of the invention for detecting fault locations in a continuous belt of glass, the starting position is fixed relative to the frame of the glass conveyor, signals indicating the distance from at least one side edge from the belt to the home position, are used to automatically control a cutter mounted further downstream along the belt path and to cut an edge portion of predetermined width from one or each side of the belt. band. Two cutting members for removing the opposite edges of the belt can be controlled by the same signals if the width of the belt remains almost unchanged. However, in most cases it will be necessary to generate independent signals indicating the transverse distances of the two side edges relative to the fixed starting position.

If the beam displacements are monitored from a fixed home position, this is preferred

* N

It may be necessary or desirable to reset the monitor periodically to maintain the reliability of the monitor signals over a period of continuous application of the method. The reset can be done manually or automatically depending on the design of the monitor used. A preferred embodiment of the invention is characterized in that signals supplied by a photo detector arranged in the home position or in a fixed position relative to this home position in response to irradiation of this photo detector by the scanning beam , automatically resets the beam displacement monitor repeatedly to ensure that it correctly represents the beam positions relative to the home position.

Preferably, the signaling means outputs (a) the distance covered by the scanning beam in successive scanning cycles from the. fixed home position to the position where the beam meets the nearest side edge of the glass, and (b) indicate the distance between the home position and a signaled error. The cross ordinate of a signaled error is hereby indicated and is equal to the difference between the signaled distances. Signaling the side edge and error positions in the form of their distances to a common starting position simplifies signal processing.

Preferably, at each scanning pass of the beam, the beam strikes a reference photodetector, thereby generating a reference signal just before the beam reaches the nearest side edge of the glass, the next beam deflection signal being processed by the signaling means signal indicating the impact of a side edge of the glass through the beam. Incidental false signals, which could be the result of beam reflections on parts of the device on which the beam falls, before it reaches the immediate vicinity of the glass, are thus disregarded. When the N.

- y - y * beam meets a side edge of the glass, the beam is deflected and this deflection can be registered by a photoelectric detector in the same way as a deviation caused by an error. If desired, the same detector can be used to detect side edges as well as errors. In any case, the generation of a reference signal makes it unnecessary for the side edge signals to be distinguished since they are distinguished from error signals, since they are always the first detector signals which are generated after a reference signal.

The above reference photodetectors can also serve as the above starting position photodetectors. However, it is better to provide the reference photodetectors and one or a pair of home position photodetectors with the 15 reference photodetectors mounted closer to the path of the glass.

It is advantageous if each scanning movement of the beam is encoded as a series of signal pulses, so that each given momentary position of the beam corresponds to a given pulse number. Such coding can be easily performed and produces signals which can be easily processed by the signaling device together with the signals indicating the irradiation of the glass edge or an error.

In a particularly favorable embodiment of the method according to the invention, there are reference photo detectors, as referred to above, in which the scanning movement of the beam is encoded as a series of signal pulses, which signal pulses are applied to a pulse counter, which converts beam positions into each scanning cycle. impulse numbers, counting from zero or other predetermined output setting, corresponding to a predetermined position of the beam outside the limits of the beam movement across the glass and irradiating the first of the two reference for each successive scan of the beam photodetectors encountering the beam causes the counter to reset to a value corresponding to the distance of the beam displacement between the starting position and that reference photodetector. This combination of features improves the reliability of the error site signals over a period of continuous application of the method.

According to the invention, a radiation beam from the same radiation source as the glass scanning beam can be generated in synchronism with the scanning of the glass for scanning a reflector having alternately reflective and non-reflective bands, the radiation guarantee, 1-0 which is the reflector equipped with bands are reflected, incident on a photoelectric detector and thereby generate the signal impulses. This monitoring of beam displacements makes it possible to generate error site signals with sufficient precision, even if there is a significant variation in the speed of the scanning beam during each scanning pass across the glass. The synchronism between the scanning of the glass and the scanning of the tape-reflector is preferably realized by using a common oscillating reflector to obtain the scanning movements of the two beams.

According to another embodiment of the invention, the positions of the scan beam in the course of each scan pass are monitored based on the time lapse from the start of the scan pass. This time course can be measured by a digital clock, which can deliver pulses, which are processed in the same way as the pulses, which are delivered in response to the radiation quanta, which are reflected by a banded reflector according to the previous embodiment. . However, when the positions of an angled oscillating beam are monitored based on time intervals, it is necessary that the amplitude of the beam oscillations be sufficiently greater than the transverse dimension of the glass to ensure that the variations in the velocity of the beam as it passes over the glass is sufficiently small to allow the required accuracy in determining the error sites. Preferably, the glass scanning beam is an angled π beam and the amplitude of the beam movement in the plane of the glass is at least twice the width of the glass. Under these conditions, the speed of the beam during the path over the glass will be sufficiently constant and the speed of movement across the glass can be taken equal to the average speed of the beam over the full arc of movement. In or near one end of the arc of movement, the scanning beam may strike a photodetector which sends a response signal to the monitoring clock thereby resetting it in order to maintain a certain accuracy for the monitoring.

The amplitude of the oscillations of an oscillating scanning beam can be automatically adjusted if the amplitude starts from a predetermined range. In the case where the beam oscillations are realized by an oscillating reflector, as described below, this automatic adjustment can be achieved by adjusting the energizing pulses for an electromagnet, which supplies the oscillating force to the reflector.

The method according to the invention can be carried out in such a way that the output signals indicate the longitudinal measurements of detected errors, ie their size measured in the direction of the transport of the glass through the scanning station. To this end, a plurality of consecutive beam passes in which error-induced signals are generated when the beam is approximately the same distance from a given edge of the glass are recorded and output signals indicating the size of the error. In the case of sheet or float glass tapes 30, any errors usually have a major dimension in the longitudinal direction of the tape and this error size is therefore usually a measure of the severity of the error as a whole.

According to a preferred embodiment of the invention, the beam is reflected back through the glass at any time during the irradiation of the glass, before reaching the photo detector (s) from which error signals originate.

- - 12 -

The method can hereby be carried out in a smaller space than would otherwise be necessary. In addition, the optical system can be more easily positioned relative to the path of the glass. For example, when examining a plate or float glass tape moving along a substantially horizontal path, the beam and the photodetector member may be positioned at a level below the tape path while the reflector reflects of the scanning beam can be arranged back through the glass above the web. Preferably, the areas of the glass irradiated at any given time by the incident and reflected scanning beams are not separated or not separated by more than two cm. In this way, the occurrence of a double signal for any given error is avoided or easily identifiable as from one error.

According to another possible, but advantageous embodiment, the radiation leaving the glass beam is split into a part, which is sent to the place of a first detector, and a part, which is sent to a second detector 20, one of these detectors being such arranged to be irradiated when the beam is deflected, while the other detector is arranged to be irradiated unless the beam is interrupted by an impermeable error or by an error of more than a predetermined degree (such as through a side edge of the glass). This joint use of two detectors, which operate in different ways, makes it possible to generate photo-detector signals which not only signal the presence of a radiation-deflecting error, but also characterize them as being of a particular type or not. In particular, this use of two detectors makes it possible to distinguish between radiating transmissive and impermeable errors.

According to an embodiment of the invention, the photodetection device detects not only deflections of the scanning beam caused by the side edges of the glass, but also faults which weaken the beam without.

% to deflect it. This method can be carried out using a single error photo detector / which is normally irradiated by the beam emanating from the glass and generates a signal only if the incident radiation falls below a predetermined minimum value because the beam is so strongly deflected that the photo detector is no longer irradiated or because the glass exerts a certain minimum attenuation influence on the beam. The minimum beam deflection at which the photo detector generates a response signal can be determined by the size of the photosensitive surface of the detector.

It is possible to realize the scanning by moving the transmitter of the scanning beam, but preferably the transmitter is stationary and the glass is scanned by a beam which is fed to the glass by an oscillating deflector, for example a reflector. deflected. In this way, the scanning movements can be effected more easily.

According to an embodiment of the invention, deviations of the beam scan paths from a predetermined plane cause irradiation of aberration photodetectors to a greater degree, the scanning path being automatically corrected by response signals from these aberration detectors. This provides useful protection against erroneous operation due to the misalignment of the scanning beam, as could be the result, for example, of moving parts of the device from preselected positions due to environmental influences.

According to another possible preferred embodiment, the glass is scanned at a frequency of at least 20 cycles per centimeter length of the glass. This better ensures that an objectionable error will be detected in a number of successive scans, which improves reliability.

The radiation beam used in the method according to the invention is preferably a laser beam. The laser beam can easily be made very narrow, which enhances the accuracy of the fault location detection.

. "X

- 14 -

The invention also relates to a device for determining the location of errors in flat glass during the course of a path, provided with means for transporting flat glass through a research station, means for generating a beam of electromagnetic radiation and repeatedly over the traversing this beam, traversing successive transverse tracks across the glass at the research station, and a photodetector for detecting when the beam strikes errors in the glass.

According to the invention, the device is characterized in that the photodetector element supplies signals indicating beam deflections or beam attenuations as well as beam deflections through the glass, the device being provided with a beam monitor for generating signals representing the beam positions. means associated with the beam monitor for recording with respect to each scanning run of the beam in one direction or periodically with respect to a succession of such scanning passes, of the position or mean position achieved by the beam in a scanning pass or scans when a first signal, caused by the beam deflection through the glass, is received from the photodetector, and a signal processor, which is designed to output in use the signals at the time of occurrence of a beam path, indicating a later signal, that the bundela indicates deflection or attenuation by the glass is received from the photodetector after the beam has left said position or average beam position and which also indicate the distance traveled by the beam in the time interval between the moment when the beam reaches said position or average position and the moment of later signal reception.

The device according to the invention can automatically signal not only the longitudinal positions of errors, but also their transverse ordinates 35 (ie their distances from a side edge of the glass), even if the side edge position varies with respect to the width of the glass conveyor. - - · during transport of the glass through the scanning station.

In the case of continuous glass belts, the accuracy of the signaled data is not affected by variations in the bandwidth passing through the glass scanning station, by waves 5 of one or both side edges of the glass relative to the direction of transport, or by "snaking" the glass on the transporter. The device is therefore particularly suitable for identifying the location of defects in newly manufactured float glass tapes.

The fact that the device can determine transverse ordinates of error positions, even if the side edges of the glass do not follow a constant line of motion, contributes to the fact that the device is equipped to withstand the radiation deflection effect which has a side edge of flat glass when use a radiation beam to and through the side edge.

The device according to the invention is intended for determining the location of errors in moving flat glass with a width (measured transverse to the direction of transport), which is smaller than the length of the track, which passes through the beam in the plane of the glass is taken. When this condition is assumed to be met, the first signal, indicating a beam deflection through the glass, which is provided by the photodetector at each pass of the beam, coincides with the incident of the beam on the nearest side edge of the glass. If a later signal is supplied by the photodetector in the same course of the beam, this may be due to the presence of an error which deflects or attenuates the radiation beam, in which case the distance traveled by the beam between the time point The pin on which the first and subsequent signals have occurred is the transverse ordinate of the error, that is, the distance to a side edge of the glass. If the later signal is caused by deflection of the beam by the other side edge of the glass, this can be disregarded by the fact that the distance which the beam is in the time interval between the reception times of the first and later signal is elapsed, almost corresponding to the width of the glass.

”> - 16 -

The photodetection device of the device according to the invention can be chosen for the detection of different types of errors, for example bubbles, grains and stones, which cause deflection of incident radiation, errors which are impermeable to the radiation used and errors in the form of discolourations, such as tin discolourations from a metal bath, which attenuate the radiation beam attenuated.

Preferably, the signal processor is connected to a glass marker or cutter control mechanism disposed along the glass transport path downstream of the scanning station, so that the marker or cutter marks the glass for error or glass cuts in places that cause the signaled error positions. In this way, a cutting member can be controlled such that the examined glass is cut into plates of desired dimensions and quality indications. The times at which error-induced signals are produced by the photodetector indicate the longitudinal position of the errors, i.e., the positions of the beam-20 traces followed across the tape in which the errors are found ; and from the times at which these signals occur, the distance between the scanning station and the marking or cutting station and the speed of the glass, the times at which these errors arrive at the marking or cutting station 25 can be determined.

The device according to the invention can alternatively be designed such that output signals are generated which indicate the longitudinal positions of errors, expressed in the distance to a transversely extending output line. For example, a device for detecting the location of errors in moving glass plates may include a photo detector which responds to the beam deflection caused by the arrival of the leading edge of a plate between the photo detector and a radiation source. In this case, the leading edge can serve as an output line and the signal processor can compare the times at which the output and error signals occur and supply output signals, the longitudinal distances of errors relative to the leading edges of the plates to hand over.

It is not necessary that the transverse ordinates of an error be determined at any time based on a glass edge plate signal (s) supplied by the photodetector at the actual beam path or passages in which the signaled error is hit. The time interval between successive beam passes across the glass is so small that at least in the case where the glass under investigation is a newly manufactured continuous band, there will be no significant change in the edge position of the glass in the detection station even over a period of time of many beam passes. For this reason, in order to simplify the signal processing means in the device according to the invention, the processing means may have the possibility of periodically determining the average of the beam positions in which photodetector signals showing the deflection of the scanning beam by a side edge of the glass are supplied for a number of consecutive beam passes, and the cross-ordinate output signals 20 are derived based on this average value.

According to a preferred embodiment of the device according to the invention, a blocking member prevents output signals from the signal processing member from indicating the location of errors which are less than a predetermined distance from a side edge of the glass. For many applications of the device according to the invention it is not important to indicate the location of errors near the side edges of the glass, because an inferior quality of the glass in the side edge parts thereof is an inevitable consequence of the glass manufacturing process itself and this edge parts are treated as filling material. The width of the unusable edges depends on the type of machine with which the flat glass is produced, on the total width of the flat glass leaving the machine and on other factors. For example, in the case of production of a flat glass belt with a width of three to four meters, each of the side edges, which is treated as waste, may have a width in the order of ten to 18-18 cm. Although the information provided by the signal processor may contain positional data even for errors that lie in the parts of the tape to be disregarded, it is better to disregard such errors. The baseline data is then limited to the more important information.

The blocking member may form part of the signal processing member and block glass error signals which should be disregarded. Alternatively, the photodetector may cooperate with a means which prevents any signals from being passed to the signal processor during any pass of the scanning beam.

According to the invention, the beam monitor 15 can generate signals indicating the beam positions with respect to a fixed starting position which is outside the limits of the beam movement across the glass. In such an arrangement, the signal processor may provide output signals indicating side edge and error locations along the scan path relative to a fixed output position. This is advantageous because such outputs are particularly suitable for automatically controlling operations on the moving glass by a device mounted a fixed distance downstream of the area where the glass is being scanned.

It is advantageous if the glass cutters are mounted in a position along the path of the glass downstream of the examination station, with means being provided which generate signals for successive scanning cycles, indicating the distances from the beam to the starting position when the beam strikes the side edges of the glass, and means are provided for controlling the cutting members in dependence on those signals, so that the cutting members cut edge portions of predetermined width of the glass during transportation from the research station . This embodiment is particularly advantageous in an installation for producing a continuous strip of flat glass, the edge parts of which have to be continuously removed because of their inferior quality. By controlling the edge cutters by signals indicating the beam displacement distances from a home position fixed relative to the glass belt conveyor, the width of the edges removed can be kept substantially constant despite changes in the lines of motion of the side edges of the tire due to tire snaking or some other cause. The cutting members for removing the opposite edges of the wall can be controlled by the same signals, if the width of the belt remains at least virtually unchanged. In most cases it will be necessary to generate independent signals indicating the transverse distances of the two side edges to the fixed starting position.

In the embodiment where the beam monitor is provided to monitor the beam movements from a fixed home position, as described above, it may be necessary or desirable for the device to include means which periodically reset the monitor to maintain the reliability of the monitoring signals over a period of continuous application of the method. Such reset means may be manually operable or automatic. According to a preferred embodiment of the invention, at least one photodetector disposed in the scanning scanning path of the beam in a fixed relationship to the home position is connected to the beam monitor so that it is automatically reset repeatedly by signals generated by the photodetector 30. supplied, ensuring that the signals from the monitor correctly indicate beam positions relative to the home position.

The signal processing means of the device according to the invention preferably supplies output signals, which (a) indicate the distance covered by the scanning beam in successive scanning cycles from the fixed starting position to the position in which the beam is the nearest side edge of the scanning beam. touches the glass and (b) indicate the distance between the starting position and the position in which the beam hits an error in the glass. Signaling the side edge and error positions, expressed in their distances from a common home position, is advantageous because of the simplification of signal processing. The signals representing distances (a) and (b) indicate the cross-ordinate of a signaled error, which is the difference between the signaled distances.

According to an advantageous embodiment of the invention. a pair of photodetectors (hereinafter referred to as reference photo-detectors) located near the opposite ends of the web, which the beam travels over the glass during movement, said reference photo-detectors connected to the signal processing means, which provides output signals which deflect the beam. through the glass, the signal processing means being so designed that it does not process a photodetector response signal to indicate a beam deflection or attenuation due to a fault in the glass, if this signal is provided during a first part of the scanning pass of the beam before the beam has reached a reference photo detector. The application of such reference photodetec. Tors has the advantage that output signal errors due to accidental reflections of radiation on the photodetectors during movement of the beam past the ends of the path across the glass are avoided.

The device preferably comprises an encoder,. that each scanning movement of the radiation beam encodes as a series of signal pulses, so that any given momentary position of the beam corresponds to a given pulse number. Such signal pulses can be easily processed along with the photodetector signals to derive the required output signals representing the error sites.

According to a preferred embodiment of the device according to the invention, there is a pulse counter for converting beam positions in each scan cycle into pulse counts> Λ - Δ I numbers, counting from zero or another predetermined output setting corresponding to a predetermined starting position of the beam, wherein the reference photodetectors are connected to the counter, so that for every 5 consecutive scanning steps of the beam, the response signal caused by irradiation of the first of the two reference photodetectors striking the beam, causes the counter to reset to a value corresponding to the beam displacement distance between the output position and the reference photodetector. This improves the reliability of the error site signals over periods of continuous use of the device, especially when the reference photodetectors are mounted close to the positions in which the beam begins to traverse the glass.

According to the invention, the device may be characterized by a reflector having alternately reflective and non-reflective bands, means which scan the reflector in a second radiation beam (hereinafter referred to as the monitoring beam), which originates from the scanning beam generator. with the scanning movements of the glass scanning beam, wherein a photodetector is arranged for irradiation by radiation quanta, which are reflected by the banded reflector, and which generates the signal pulses in response to this irradiation. This embodiment for encoding the scanning movements of the scanning beam has the advantage that the accuracy with which the pulse signal represents the position of the beam at a given time is not affected by fluctuations in the speed of the beam during the scanning movement over the glass. Such velocity fluctuations are inevitable in the event that the scanning beam scans the glass with an reciprocal angular movement and these fluctuations will be of a significant magnitude unless the angular movement of the beam is much greater than the width of the glass.

Accuracy and reliability are enhanced by deriving the monitoring beam from the same source - 22 - as the scanning beam. Preferably, the synchronism of the scanning and monitoring beams is ensured by a common oscillating deflector (eg a reflector) used to obtain the scanning oscillations of the two beams.

According to an embodiment of the invention, the device may be characterized by a digital clock, which monitors the movement of the scanning beam according to the passage of time from the beginning of each scanning pass and provides 10 pulses indicating the beam position. Such a beam monitor avoids the need to divert a second radiation beam from the scan beam generator, nor does it require the tape reflector and associated photodetector. It is possible to obtain results with the same accuracy by means of a monitoring clock. However, if the scanning beam oscillates at an angle (as opposed to rotating in one direction, so that the path across the glass is followed in the same direction each time), it is necessary that the amplitude of the beam movements be considerably greater than the width of the beam. glass. Preferably, these amplitudes are at least twice the width of the glass. If a home position photodetector is used, which is preferable, this photo detector is preferably connected to the clock so that it is reset to an output setting every time the home position photo detector is irradiated by the scanning beam.

According to a preferred embodiment, the signal processor provides output signals during operation which indicate the number of consecutive beam passes in which a beam deflection or attenuation signal is supplied when the beam is approximately the same distance from a side edge of the glass. Such an embodiment of the signal processing device has the advantage that the output signals of the device provide not only information about the location of radiation-deflecting errors, but also about the dimensions of the errors. The signals, which the error dimensions' ^ ’Λ v

, ·.! J

- 23 - can be further processed by a selection mechanism, which determines the quality of different areas of the flat glass by the number, distribution and dimensions of errors therein.

According to another possible, yet advantageous embodiment, a reflector is arranged to reflect the scanning beam back through the glass, the photo-detector means for detecting beam deflections or attenuations caused by errors such that this responds to deflections or attenuations of the reflected beam after it comes out of the glass a second time. This has the advantage that the device takes up less space. According to a possible embodiment, the radiation beam source and the photodetector 15 for detecting error-induced beam deflections or attenuations may be installed below the glass transport path and the scanning beam reflector may be positioned above this path. The beam source may be arranged to send the beam directly upward to the path of the glass, or the beam may be directed horizontally or substantially horizontally toward a reflector which reflects the beam upwardly. This latter embodiment allows the device to operate under conditions where the optical length of the scanning system is greater than can be accommodated in the available vertical space. The beam source may, of course, alternatively be arranged such that the scanning beam is directed downwardly onto the glass, but this arrangement entails more problems and costs for mounting the device in the use position.

According to another possible, but nevertheless particularly advantageous embodiment, the device is characterized by a beam splitter for splitting the scanning beam after it leaves the path of movement of the glass into two derived beams, wherein separate photodetectors for the derived beams are arranged so that during operation one of the photodetectors - 24 - - \ is irradiated only when the scanning beam is deflected through the glass, while the others will be irradiated unless the scanning beam is prevented from being transmitted through the glass or more is then deflected to a predetermined degree. The advantage of this embodiment with two detectors is that information signals can be generated, which not only indicate the location of an error, but also whether or not the error falls under a certain type of category. For example, if one detector detects a beam deflection, but the other detector remains unexposed, this indicates a type of error that has only a minor refractive effect on the incident radiation; and if none of the detectors are irradiated at a given time, this indicates that the scanning beam is intercepted by an impermeable zone in the glass.

According to the invention, the photodetection means comprises a photodetector, which is arranged such that during operation the photodetector is normally irradiated by the beam emerging from the glass, the photodetector providing a signal if the incident radiation falls below a predetermined minimum threshold value. coming. Such a photodetector can detect and signal errors that attenuate the scanning beam without deflecting it. The photodetection device of the device should, of course, also detect beam deflections as caused by the edges of the glass. Such large beam deflections as well as such beam attenuations can be detected by one and the same photodetector, which has a sensitive surface of such a size that it remains irradiated unless the scanning beam is deflected by more than a predetermined minimum.

Preferably, the beam generator is held so that it remains stationary during use of the device, while the device includes a deflector, for example, a reflector, coupled to an oscillation means for oscillating the deflector for obtaining the scanning movements of the beam. A particularly recommended oscillation mechanism for oscillating the deflector includes a torsion member attached; -1 - 25 - * to the deflector, and an electromagnetic or other means for oscillating this element at the natural frequency. It has been found that such an oscillation system works accurately and reliably. The oscillating mass can be very small. The electromagnet can be energized by voltage pulses, which are controlled by the movement of the deflector itself, so that the pulse frequency coincides with the ground resonance frequency of the torsion element.

According to an embodiment of the invention, aberration photodetectors provide a response signal, if the trajectory of the scanning beam deviates from a predetermined plane by more than a predetermined amount, and they are connected to an adjustment means, whereby an automatic correction adjustment of the trajectory is reached. This provides useful protection against malfunctioning due to misalignment of the scanning beam, which may result, for example, when parts of the device move from preset positions. due to environmental influences.

It is preferable if the means for generating the scanning beam, the error photo detectors and the means for obtaining the scanning movements of the beam are integrated into one unit. This facilitates installation of the device in a suitable position relative to a particular glass conveyor.

Preferably, the radiation beam transmitter is a laser. The use of a laser has the advantage that the currently irradiated area of the glass is as small as possible, so that the device has a particularly high resolution.

The invention is explained in more detail below with reference to the drawing, in which an exemplary embodiment is shown.

Fig. 1 schematically shows a scanning and detection system of the device according to the invention.

FIG. 2 is a block diagram of the signaling device of the invention.

In Fig. 1 the scanning and detection system is shown.

S

Given in the direction of transport of a glass belt 1.

A beam transmitter 2 and a number of photodetectors for monitoring the scanning movements of the beam and detecting beam deflections are mounted in a location below the path of the glass tape, while above this path a concave reflector 3 is arranged for reflecting back to the detectors of radiation.

The beam emitter 2 preferably consists of a laser gun and the radiation beam will hereinafter be referred to as laser beam 10. The transmitter 2 is arranged in a fixed position. The laser beam is split by a half mirror reflector 4 into a transmitted part 5 and a reflected part 6. This reflected part is further reflected by reflectors 7 and 8. The transmitted beam 5 and the beam 9 reflected by the mirror 8 both fall on a reflector 10. The center of the reflector 10 lies in the bisectric plane of and contains the axis of curvature of the reflector 3.

The reflector 10 is mounted on a torsion bar and is oscillated by an electromagnet, the excitation circuit of which contains a switch, which is operated by the reflector movements, so that the reflector oscillates at the natural frequency of the bar. The oscillation frequency of the reflector 10 is 800 cs / s. The oscillation of the reflector 10 causes the reflected beam 11 to scan the glass ribbon 1 at this frequency. If it is assumed that the speed of the glass ribbon is 20 cm. per second, this means that the belt undergoes forty scanning cycles through the bundle per centimeter of length.

The beam 9 is converted by the oscillating reflector 10 into an oscillating reflected beam 12, which swings back and forth precisely synchronously with the scanning beam 11. The beam 12 scans a concave striped reflector 13, which includes reflective and non-reflective strips, which alternate along the path followed by the beam.

The reflector 13 therefore intermittently reflects the incident light, the light being reflected as a series of discrete quanta or impulses, such as through the lower ones.

broken line 14 is indicated. These light pulses are directed to the location of a photo detector 15. The center of this photo detector and the center of the oscillating reflector 10 are arranged symmetrically with respect to the bisector plane of the reflector 13, which also contains the axis of curvature of the reflector 13, so that all light pulses are received by the photo detector 15 despite the oscillatory movement of the beam 12, whereby the pulses are generated. The synchronous scanning beam 12, the striped reflector 13 and the detector 15 together serve to encode the scanning movements of the main scanning beam 11 as a series of electrical signal pulses which are processed by a signal processor in the manner described below.

The oscillation amplitude of the oscillating reflector 10 is such that the path traveled by the beam in the plane of the glass tape extends considerably beyond the limits of the tape path. In fact, the amplitude is such that the beam moves from a starting position indicated by line X on one side of the track to a position indicated by line Y on the other side of this track. Consequently, the bundle moves through both side edges of the belt during each transverse scanning movement of the bundle. In the illustrated exemplary embodiment, reference photodetectors 16 and 17 are disposed near the opposite ends of the reflector 3 for a purpose yet to be described and the scanning beam 11 also travels over these photodetectors with each scanning movement.

As it moves across the glass tape, the scanning beam 11 is reflected back through the tape by the reflector 3 as a reflected scanning beam 18. The optical arrangement of the system is such that the locations where the glass is currently irradiated by the incident and reflected scanning beams, very close to each other (preferably within two centimeters of each other) and preferably overlapping or adjacent to each other.

The reflected beam 18 falls upon leaving the glass ribbon on a semipermeable reflector 19 and is thereby split into a transmitted portion 20 and a reflected portion 21. In the drawing, it is inevitable that the incident and reflected scanning beams 11 and 18 both lie in the same plane, but in reality this is not the case otherwise the oscillating reflector 10 and the curved reflector 13 would block the reflected scanning beam. The transmitted portion 20 goes to a photodetector 22, while the reflected portion 21 goes to a photodetector 23. An intercept disk 24 is arranged centrally in front of the sensitive surface of the photodetector 22 and intercepts the transmitted portion 20 of the reflected scanning beam 18, if and for as long as the scanning beam 11 and therefore also the reflected scanning beam 18 are not deflected by the glass tape. Deflection of these beams occurs when the scanning beam meets a side edge of the belt and can also occur when the beams encounter an error in the glass. Any significant deflection of the scanning beam results in the photo detector 22 being irradiated by the transmitted portion 20 of the reflected scanning beam.

The reflected portion 21 of the reflected scanning beam 18 always strikes the photosensitive surface of the photodetector 23, except when the reflected scanning beam is deflected through the glass. If desired, the photodetector 23 may have a sensitivity such that it responds to a decrease in the incoming radiation current below a certain threshold value, so that the photodetector can signal when the scanning beam encounters glass defects which attenuate the beam without causing deflection.

The successive distance increments, which are covered by the beam along the reflector 3 and the striped reflector 13 at any given scanning pass, do not have a constant relationship to the distance increments which are simultaneously deposited on the glass. In order to minimize the signaling error due to this inequality, it is desirable that the longest possible optical path is present between the oscillating reflector 10.

"N

- 29 - on the one hand, and the glass and reflector 3 on the other. Good results have been achieved with a distance between the oscillating reflector and the reflector 3 of 7-8 meters.

In the block diagram according to Fig. 2 of the signaling element, the photo detectors 15, 16, 17, 22 and 23 of Fig. 1 are shown on the left. The beam monitoring pulses supplied by the photodetector 15 are applied after amplification to a pulse counter 26, as are the signals supplied by the photodetectors 16, 17 mounted close to the reflector 3 for the scanning beam.

The monitoring pulses supplied by the photodetector 15 during a beam movement to the right according to the view of Fig. 1 are recorded as consecutive increases in the value recorded by the counter, while monitoring pulses delivered during a scan to the left is recorded as successive decreases in this value. The counter can be reset periodically so that the pulse count continues to represent the beam position 20 relative to the fixed starting position X with sufficient accuracy. Such a reset can be done manually at the required intervals. For example, the counter behavior can be monitored with a display, which shows when a reset is needed.

Preferably, however, the counter is automatically set and reset. Preferably, the reset occurs automatically in each cycle of the scanning beam or at an interval of two or more cycles, in response to the incident of the beam on one or more photodetectors. For example, if a home position photodetector is arranged in X, the counter can be automatically reset to a home position, eg zero, by a signal sent by this photodetector in response to its irradiation. In this case, it is preferable to reset the counter in response to the irradiation of the reference photodetectar 16 during a movement to the right of the beam and to reset again in response to the irradiation of the reference photodetector. 17 during a leftward movement of the beam so that the pulse counts recorded when the beam is at these reference photodetectors properly represent the respective distances of these photodetectors from the home position X. As an alternative, the home position photo detector can be omitted and the counter reset can only be caused by signals from the reference photo detectors 16 and 17, which signals serve to set the counter to pulse counts, which are at the correct the aforementioned distances.

Due to the described counter reset, the value recorded by the counter 26 is always a correct representation of the position of the beam relative to the fixed starting position.

Monitoring pulses, which indicate the varying position of the scanning beam, are applied from the counter 26 to an edge position computer 27, to which the photo detectors 22, 23 are connected via an OR circuit 32. The computer 20 includes a gate circuit 28, a number adjuster 29, an accumulator / calculator 30 and a register 31.

During each scanning movement of the scanning beam 11, the first deflection of the beam through the glass occurs when the beam first strikes a side edge of the glass. The photodetector signal caused by this is applied to the associated gate circuit 28, as well as the beam position monitoring pulses of the counter 26. The gate "circuit passes a counter output from the counter 26 to the accumulator / calculator 30, which signal indicates the beam-30 position at the moment when the beam deflection signal is supplied to the gate circuit through the OR circuit 32.

Such transmission of the signal through the gate circuit 28 takes place in each of a number of scanning cycles of the beam, this number being determined by the setting of the number adjuster 29. The beam position signals transmitted through the gate circuit in the predetermined set number of scan cycles are accumulated in the accumulator V. 31, s, tor / arithmetic unit 30 and this unit then calculates the averages of the accumulated beam positions and sends signals indicating these averages to the edge position register 31. According to one embodiment, the set number is 128 cycles and adding the 128 guard pulse numbers, representing the location of a band edge, and calculating the average value takes 8 seconds.

The signals representing the average positions of the respective glass edges over the time intervals determined by the setting of the number adjuster 29 are applied to a comparator circuit. 33. They are also supplied to a mechanism for controlling the positions of glass cutters for cutting the glass ribbon at a location downstream of the error detection station, which includes cutting members that position the glass ribbon in plates of the desired size and quality indication cutting, as well as cutting means for continuously removing the side edges of the tire. By using said signals to control the movement of the edge cutters in a direction transverse to the belt path, the width of the edges cut from the glass can be kept constant even during periods, wherein a variation in the positions of the tape edges occurs as it passes through the cutting station. The glass edge position signals can optionally also be applied to a display or recording device 34, which can be inspected by a operator.

The beam positions, which coincide with beam deflections due to errors in the glass, are registered by an error register 35. To this end, signals from the photodetectors 22 and 23 are applied to this error register 35 via a blocking circuit 36, one of the purposes of which is to noise filtering, including false signals, is caused by incidental light reflections from parts of the glass transport and research installation. The register 35 receives beam position monitoring signals from the counter 26 and sends signals to the comparator circuit 33 indicating the positions of the beam at times when beam deflection signals are received from the photo detectors 22 and 23. The blocking circuit 36 also receives signals from the photodetectors 16 and 17, which are arranged close to the ends of the scanning beam reflector 3 and only transmit deflection signals to the error register 35, which occur during the movement of the beam between these detectors.

Signals indicating beam positions coinciding with errors signaled by detectors 22 and 23, such as the band edge position signals from the edge position register 31, are applied to the comparator circuit 33 and a gate circuit 37. In the comparator circuit, the error position signals of the error register 35 compared to the edge position signals of register 31 to determine which errors are more than a predetermined distance from each edge of the tape. In response to error position signals falling into this category, the comparator circuit sends a release signal to the gate circuit 37. This signal opens the gate circuit and passes the corresponding error position signals from the error register 35 to a microprocessor 38. This microprocessor is via an OR circuit 39 is also connected to the reference detectors 16, 17, so that signals are received, indicating the moments when the beam starts a new pass over the glass. These signals are of course important for determining the dimensions of the individual errors.

The microprocessor has three output lines 40-42.

Line 40 provides signals representing the transverse distance of the errors from the home position X. The line 4130 provides signals indicating the nature of a signaled error. Different categories of errors are indicated depending on whether or not they cause a beam deflection sufficient to cause interruption of the irradiation of the detector 23. Line 42 35 provides signals indicating the length of a signaled error, a factor which can be deduced from the number of consecutive scans of the beam in which a beam is 33.

Deflection is signaled at approximately the same beam position. In this connection, it is often advantageous to record small errors, which are very close together, as a single error. For example, if error signals in the same position occur more than once within an interval of up to forty scan cycles, such signals can be treated as representing a single error.

The signals, which are sent via lines 40-42, are sent more or less simultaneously with the incident beam incident on the errors which cause these signals. The times at which these signals are violated therefore indicate the longitudinal positions of the errors. Taking into account the speed of movement. of the glass, these times at which error signals are sent provide the ability to determine the arrival times of the errors in any position downstream of the scanning station.

For classifying the size, the microprocessor is provided with a number of multi-cell registers. Error position signals are supplied to a cell of each register depending on the cross-ordinate of the signaled position. Error position signals representing corresponding cross-ordinates are supplied to the same register cells. The cells of each register have 25 associated output gates to which a threshold charge value is provided that is appropriate for a particular category of error size. There are several threshold load values, one for each register. The number of registers corresponds to the number of different size categories in which errors must be classified. Position signals corresponding to the same cross-ordinate and sent in direct successive passes of the scanning beam have a cumulative charge effect on the cells to which they are supplied. The number of charge increments accumulated in each particular cell therefore indicates the longitudinal size of the error causing the charging of the cell. Adding a charge increment to each - 34 - determined cell is accompanied by a decrease in the threshold charge value, which is delivered to the associated port. If the threshold charge value on a gate falls below zero, a signal is sent through the associated cell, indicating that the longitudinal dimension of the error having the cross ordinate associated with that cell is at least equal to the threshold value which is represented by the preset gate load. If a charge increment is not added to a cell at a given scan 10 of the beam, a charge previously accumulated by the cell and the remaining charge increment of the associated cell are automatically discharged.

A favorable refinement of the described device comprises installing photodetectors, which are installed side by side on the correct scanning path of the scanning beam over the reflector 3, which photodetectors are connected to an adjustment means for the oscillating reflector 10, so that the reflector is automatically adjusted if vibration or other disturbing influence causes the beam to deviate from the correct path across reflector 3 in lateral direction 20. Such additional photodetectors can for instance be arranged near one end of the reflector 3, for instance in a plane perpendicular to the plane of Fig. 1 and between the reference photodetector 16 and the associated end of the reflector 3.

According to the invention, the photo detector 22 can be omitted. Only the photo detector 23 is then used to detect the glass edge positions and the error positions in the glass. In this case, the photo detector 23 can operate in the manner described or it can be designed in such a way that it also supplies a signal when the incident amount of radiation falls below a certain value, as a result of which the beam is weakened, which is caused, for example, by a discoloration on the glass is caused, is indicated. Under these circumstances, the device will not signal the location of errors, which have only a small deflective effect on the scanning beam. In the case where a single glass edge 35 and error position photodetector is used, the semipermeable mirror 19 is of course unnecessary. The photo detector may be in the position of the photo detector 22 in the drawing. Various other arrangements of photo detectors are of course possible within the scope of the invention.

For example, the photo detector 23 can remain in the drawn position for detecting large beam deflections and the photo detector 22 can be used without the mask 24 for detecting beam attenuations.

J

Claims (43)

1. A method of determining the location of errors in flat glass during a trajectory, wherein the glass is scanned with a beam of electromagnetic radiation and a photodetector is used, whereby the incident of the beam is detected for errors in the glass characterized in that - the glass is scanned by a beam of electromagnetic radiation moving across the path of the glass, so that the beam follows successive transverse tracks across the glass and at each scanning pass the beam traverses the opposite side edges of the glass goes; - the photodetector means for detecting deflections of the beam caused by hitting one or each side edge of the glass and for detecting attenuation or deflections of the beam due to errors in the glass; - which photodetector member forms part of a signaling device which supplies output signals, which indicate positions along the scanned length of the glass according to the transverse tracks, in which errors are detected and indicate the distances measured along these tracks between these errors and a side edge of the glass, and - wherein the output signals are used to identify the signaled error sites. 2. Method according to claim 1, characterized in that the signaling device generates output signals which are used as a control factor in automatically controlling a glass marking unit or cutter for marking or cutting the moving glass at a location downstream of the station , where the glass is scanned. Method according to claim 1 or 2, characterized in that an average value is derived from the glass edge deflection signals supplied by the photodetector member at a number of consecutive 'N - 37' passes of the beam over the glass. in one direction, which average value is used in deriving the output signals indicating the error locations. Method according to any one of the preceding claims, characterized in that the output signals are only generated for errors which are further than a predetermined distance from each side edge of the glass. Method according to any one of the preceding claims, characterized in that the displacement of the beam is monitored by a monitoring device which supplies signals indicating the beam positions relative to a fixed starting position, which are outside the boundaries of the beam movement over the glass, which signals are used by the signaling means to derive the output signals. Method according to claim 5, used for detecting fault locations in a continuous band of glass, characterized in that the starting position is fixed relative to the frame of the glass conveyor, wherein 20 signals indicate the distance from at least one side edge from the belt to the home position, are used to automatically control a cutter mounted further downstream along the belt path and to cut an edge portion of predetermined width from one or each side of the belt. band. 7. Method according to claim 5 or 6, characterized in that signals supplied by a photodetector, which is arranged in the starting position or in a fixed position relative to this starting position, in response to irradiation of this photodetector by the scanning beam, automatically resets the beam displacement monitor repeatedly to ensure that it correctly represents the positions of the beam relative to the home position. A method according to any one of claims 5-7, characterized in that the signaling device supplies output signals (a) the distance covered by the N-38 scanning beam in successive scanning cycles from the fixed starting position to the position , in which the beam meets the nearest side edge of the glass, and (b) indicate the distance between the starting position and a signaled error. 9. A method according to any one of the preceding claims, characterized in that with each scanning run of the beam, the beam hits a reference photo-detector, whereby a reference signal is generated just before the beam reaches the nearest side edge of the glass, wherein the beam next beam deflection signal by the signaling means is processed as a signal indicating the impact of a side edge of the glass through the beam. 10. A method according to any one of the preceding claims, characterized in that each scanning movement of the beam is encoded as a series of signal pulses, so that any given momentary position of the beam corresponds to a given pulse number. 11. Method according to claims 9 and 10, characterized in that the signal pulses are applied to a pulse counter, which converts beam positions into pulse numbers in every scanning cycle, counting from zero or another predetermined output setting corresponding to a predetermined position of the beam outside the limits of the beam movement across the glass, and for each successive beam scan, irradiating the first of the two reference photodetectors encountering the beam, resetting the counter to a value which corresponds to the distance of the beam displacement between the home position and that reference photode detector. 12. A method according to claim 10 or 11, characterized in that a radiation beam originating from the same radiation source as the glass scanning beam is generated in synchronism with the scanning of the glass for scanning a reflector, which alternately reflects and non-reflective bands the radiation quanta reflected by the banded reflector - jy - incident on a photoelectric detector and thereby generate the signal pulses. Method according to claim 10 or 11, characterized in that the positions of the scanning beam 5 in the course of each scanning pass are monitored on the basis of the time lapse from the start of the scanning pass, Method according to claim 13, characterized in that the scanning beam is an angle-oscillating beam, wherein the amplitude of the beam movement 10 in the plane of the glass is at least twice the width of the glass. 15. A method according to any one of the preceding claims, characterized in that the number of consecutive beam passes in which error-induced signals are generated when the beam is at approximately the same distance from a certain edge of the glass , are recorded and produce output signals indicating the size of the error. Method according to any one of the preceding claims, characterized in that the beam is reflected back through the glass at any time during the irradiation of the glass, before the photodetector (s) from which error signals originate are reached. Method according to any one of the preceding claims, characterized in that the radiation beam leaving the glass is split into a part which is sent to the site of a first detector and a part which is sent to a second detector one of these detectors being arranged to be irradiated when the beam is deflected, while the other detector is arranged to be irradiated unless the beam is interrupted by an impermeable error or by an error in is deflected more than a predetermined degree. A method according to any one of the preceding claims, characterized in that the photodetection device not only detects deflections of the scanning beam as caused by the side edges of the glass, but also detects 40 errors, which weaken the bundle without deflecting it. 19. A method according to any one of the preceding claims, characterized in that the scanning beam is emitted by a stationary transmitter and the scanning beam is deflected towards the glass by an oscillating deflector. 20. A method according to any one of the preceding claims, characterized in that deviations of the scanning paths of the beam from a predetermined plane cause irradiation of deviation photodetectors to a greater extent, the scanning path being automatically corrected by response signals from these deviation detectors. . 21. A method according to any one of the preceding claims, characterized in that the glass is scanned at a frequency of at least twenty cycles per centimeter of length of the glass. 22. Method according to any one of the preceding claims, characterized in that the scanning beam is a laser beam. 23. Device for determining the location of errors in flat glass during the course of a path, provided with means for transporting flat glass through a research station, means for generating a beam of electromagnetic radiation and making it pass repeatedly of this beam, through which successive transverse tracks are traversed across the glass in the examination station, and a photodetector for detecting when the beam strikes errors in the glass, characterized in that the photodetector delivers signals which cause beam deflections or beam attenuations as well as beam deflections through the beam. glass, wherein the device is provided with a beam monitor for generating signals representing the beam positions, the beam monitor includes means for registering each direction of scanning of the beam in one direction or periodically with respect to a succession of such a bag corridors, from the - .41 -. position or average position achieved by the beam in a scan or scan pass when a first signal caused by the beam deflection through the glass is received from the photodetector, and a signal processor designed that, during use, output signals are provided indicating the time of occurrence of a beam pass in which a later signal indicating the beam deflection or attenuation through the glass is received from the photodetector after the beam has said position or average beam positions which also indicate the distance traveled by the beam in the time interval between the moment at which the beam reaches said position or average position and the moment of the later signal reception. 24. Device as claimed in claim 23, characterized in that the signal processing member is connected to a control mechanism of a glass marking or cutting member, which is arranged along the transport path of the glass downstream of the scanning station, so that it is marking or cutting. The device marks the glass in places of error or cuts the glass in places that cause the error positions identified. 25. Device as claimed in claim 23 or 24, characterized by a blocking member which prevents 25 signals from the signal processing member indicating the location of errors which are less than a predetermined distance from a side edge of the glass. 26. An apparatus according to any one of claims 23-25, characterized in that the beam monitor 30 generates signals indicating the beam positions relative to a fixed starting position which is outside the limits of the beam movement across the glass. 27. An apparatus according to claim 26, characterized in that glass cutters are mounted at a position along the path of the glass downstream of the examination station, wherein means are provided which generate signals for successive scanning cycles, which distances from A - indicate the beam to the starting position, when the beam strikes the side edges of the glass, and means are provided for controlling the cutting elements in dependence on those signals, so that the cutting elements have edge parts with a predetermined width of cut the glass during transport from the research station. 28. Device according to claim 26 or 27, characterized in that at least one photodetector, which is arranged in the scanning scanning path of the beam in a fixed ratio to the starting position, is connected to the beam monitor, so that it is automatically is repeatedly reset by signals supplied by the photodetector, ensuring that the signals from the monitor correctly indicate the beam positions relative to the home position. 29. Device according to any one of claims 26-28, characterized in that the signal processing device supplies output signals which (a) indicate the distance covered by the scanning beam in successive scanning cycles from the fixed starting position to the position in which the beam touches the nearest side edge of the glass and (b) indicates the distance between the home position and the position where the beam touches an error in the glass. 30. Device according to any one of claims 23-29, characterized by a pair of photodetectors (hereinafter referred to as reference photodetectors) located near the opposite ends of the web passing through the beam during movement over the glass which reference photodetectors are connected to the signal processor, which provides output signals indicating the beam deflection through the glass, the signal processor being such that it does not process a photodetector response signal as an indication for a beam deflection or attenuation due to an error in the glass, if this signal is supplied during a first part of the beam scan before the beam has reached a reference photodetector. - 43 -; 31. An apparatus according to any one of claims 23-30, characterized by an encoder, which encodes each scanning movement of the radiation beam as a series of signal pulses, so that any given momentary position of the beam 5 corresponds to a given pulse number. 32. An apparatus according to claims 30 and 31, characterized by a pulse counter for converting beam positions in each scanning cycle into pulse numbers, counting from zero or another predetermined output setting corresponding to a predetermined beam output position, wherein the reference photodetectors are connected to the counter so that for each successive scan of the beam, the response signal caused by irradiation of the first of the two reference photodetectors striking the beam resets the counter caused to a value corresponding to the beam displacement distance between the home position and the reference photodetector. 33. Device as claimed in claim 31 or 32, characterized by a reflector with alternately reflecting and non-reflecting bands, means which allow the reflector to emit a second radiation beam (hereinafter referred to as the monitoring beam) originating from the scanning beam generator. scanning in synchronism with the scanning movements of the glass scanning beam, wherein a photodetector is arranged for irradiation by radiation quanta, which are reflected by the banded reflector, and which generate the signal pulses in response to this irradiation. 34. An apparatus according to claim 31 or 32, characterized by a digital clock, which monitors the movement of the scanning beam according to the passage of time from the beginning of each scanning pass and provides signal pulses indicating the beam position. A device according to any one of claims 23-34, 35, characterized in that the signal processing device supplies during operation output signals which indicate the number of consecutive beam passes in which a beam deflection or attenuation signal is <- 44 - f '** is supplied when the bundle is approximately the same distance from a side edge of the glass. 36. An apparatus according to any one of claims 23-35, characterized in that a reflector is arranged so that it reflects the scanning beam back through the glass, the photodetector means for detecting beam deflections or attenuations caused by errors being so arranged that it responds to deflections or attenuations of the reflected beam after it comes out of the glass a second time. 37. Apparatus according to any one of claims 23-36, characterized by a beam splitter for splitting the scanning beam after it leaves the path of movement of the glass into two derived beams, wherein separate photodetectors for the derived beams are arranged, so arranged , that during operation one of the photodetectors is irradiated only when the scanning beam is deflected through the glass, while the others will be irradiated unless the scanning beam is prevented from being transmitted through the glass or is deflected more than predetermined. 38. Device according to any one of claims 23-37, characterized in that the photo-detecting element comprises a photo-detector, which is arranged such that during operation the photo-detector is normally irradiated by the beam emerging from the glass, wherein the photo-detector is provides a signal if the incident radiation falls below a predetermined minimum threshold value. 30 39. An apparatus according to any one of claims 23-38, characterized in that the beam generating means is held stationary, while a deflection means is provided which is coupled to an oscillation means for oscillating the deflection means, thereby causing the scanning movement of the beam 35 is acquired. 40. Device as claimed in claim 39, characterized in that the oscillation member for the deflector member comprises a torsion element / which is attached to the deflector member, and an electromagnetic or other member for oscillating this element on the natural frequency. 41. An apparatus according to any one of claims 23-40,5, characterized by aberration photodetectors which provide a response signal if the scanning beam trajectory deviates more than a predetermined amount from a predetermined plane, and which are connected to an adjustment means, whereby an automatic correction setting of the track is reached. 42. An apparatus according to any one of claims 23-41, characterized in that the means for generating the scanning beam, the error photodetectors and the means for obtaining the scanning movements of the beam are integrated in one piece. 43. An apparatus according to any one of claims 23-42, characterized in that the means for generating the beam is a laser. * /