EP1210586A1 - Detection of inclusions in glass - Google Patents

Detection of inclusions in glass

Info

Publication number
EP1210586A1
EP1210586A1 EP00958046A EP00958046A EP1210586A1 EP 1210586 A1 EP1210586 A1 EP 1210586A1 EP 00958046 A EP00958046 A EP 00958046A EP 00958046 A EP00958046 A EP 00958046A EP 1210586 A1 EP1210586 A1 EP 1210586A1
Authority
EP
European Patent Office
Prior art keywords
inclusion
glass
detected
inclusions
radiation
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
EP00958046A
Other languages
German (de)
French (fr)
Other versions
EP1210586A4 (en
Inventor
Geoff Edwards
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Resolve Engineering Pty Ltd
Original Assignee
Resolve Engineering Pty Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Resolve Engineering Pty Ltd filed Critical Resolve Engineering Pty Ltd
Publication of EP1210586A1 publication Critical patent/EP1210586A1/en
Publication of EP1210586A4 publication Critical patent/EP1210586A4/en
Withdrawn legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84Systems specially adapted for particular applications
    • G01N21/88Investigating the presence of flaws or contamination
    • G01N21/89Investigating the presence of flaws or contamination in moving material, e.g. running paper or textiles
    • G01N21/892Investigating the presence of flaws or contamination in moving material, e.g. running paper or textiles characterised by the flaw, defect or object feature examined
    • G01N21/896Optical defects in or on transparent materials, e.g. distortion, surface flaws in conveyed flat sheet or rod
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84Systems specially adapted for particular applications
    • G01N21/88Investigating the presence of flaws or contamination
    • G01N21/95Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
    • G01N21/958Inspecting transparent materials or objects, e.g. windscreens

Definitions

  • This invention relates to a method and apparatus for detecting inclusions in glass plate, particularly nickel sulphide (NiS) inclusions.
  • the invention further relates to an apparatus and method for classifying the detected inclusions.
  • NiS inclusions have been known for many years. In particular, it has been known for the last thirty years that heat-strengthened or fully toughened glass plates that have NiS inclusions are subject to spontaneous failure. In short, the presence of NiS inclusions can result in shattering of the glass plate. Unfortunately, these inclusions have proven to be extremely difficult to exclude from the manufacturing process. They are also very difficult to detect. This has led to potentially unsafe glass being used in buildings. The detection and classification of NiS inclusions has proven to be a very difficult problem. A heat-soaking process has been developed in which the manufactured glass is maintained at an elevated temperature. The theory is that almost all glass which contains inclusions that could cause spontaneous failure will fail during the heat soak process.
  • Reflections from the front and back surface of the glass produce two lines in the image that define the boundaries within which an inclusion may be located. Luminous points situated between the two lines are detected as inclusions within the glass.
  • the apparatus does not identify the nature of the inclusion nor is the apparatus suited to scanning of glass sheets in situ on a building.
  • a further apparatus for detecting flaws in glass is described in United States patent number 4597665, assigned to Tencor Instruments. As in the Flachglas device, the Tencor apparatus measures reflected laser light above and below the plane of the glass sheet. In most applications it is impractical to position detectors on both sides of the glass being tested.
  • the invention resides in an apparatus for detecting inclusions in glass comprising: one or more lasers emitting one or more coherent beams of radiation directed at the glass at one or more known angles; means for detecting radiation scattered by inclusions within the glass; means for scanning the coherent beams across the glass; and means for recording the location of a scatte ⁇ ng inclusion.
  • the apparatus may further comprise means for classifying the inclusions.
  • the means for classifying the inclusion suitably comprises a camera that records a pattern of laser radiation scattered from the detected inclusion.
  • the means for classifying may also include means for discriminating a pattern that indicates an inclusion is not NiS
  • the means for classifying may further comprise categorising means for positively categorising the detected inclusions that are NiS.
  • the categorising means is suitably a spectroscopic means that analyses radiation scattered from the inclusion and categorises the scattered radiation by a spectroscopic signature.
  • the invention resides in a method of detecting inclusions in glass including the steps of: directing laser radiation at the surface of the glass; scanning the laser radiation over the glass; detecting radiation scattered from inclusions in the glass; and recording the coordinates of the inclusions.
  • the method further includes the steps of detecting first detected radiation scattered from a first beam of laser radiation, detecting second detected radiation scattered from a second beam of laser radiation, and only recording the coordinates of an inclusion if the first detected radiation and the second detected radiation are separated in a known way.
  • the second beam of laser radiation may be a reflection of the first beam.
  • the method preferably includes the further steps of: viewing the detected inclusion with a camera; and classifying the inclusion according to the pattern of laser radiation scattered from the inclusion.
  • the step of classifying classifies the inclusions that are not NiS.
  • the method may also include the further steps of: spectroscopically analysing the radiation scattered from the detected inclusions; and categorising the identified inclusions as NiS by a spectroscopic signature.
  • FIG 1 is a sketch of an apparatus for detecting inclusions in glass
  • FIG 2 is a schematic side view of a first embodiment of a detection unit
  • FIG 3 is a schematic side view of a second embodiment of a detection unit
  • FIG 4 is a schematic side view of a classification unit
  • FIG 5 is a sketch showing the pattern of radiation scattered from a smooth inclusion
  • FIG 6 is a sketch showing the pattern of radiation scattered from a rough inclusion
  • FIG 7 is a sketch showing another view of the pattern of radiation scattered from a smooth inclusion
  • FIG 8 is a sketch of a detector head incorporating the detection units of FIG 2
  • FIG 9 is a sketch of a detector head incorporating the detection units of FIG 3;
  • FIG 10 is a timing diagram showing the method of determining size and depth of an inclusion.
  • FIG 1 there is shown a sketch of an apparatus 1 for detecting inclusions in glass sheets 2.
  • the apparatus 1 includes a detector head 3 that can be scanned over the area of the glass sheet 2.
  • the detector head 3 incorporates multiple detection units, such as 4.
  • the glass sheet 2 may be scanned in situ on a building or on a production line.
  • the mechanical frame 5 of the apparatus is made from extruded aluminium for strength and light weight.
  • the frame is made to a suitable size to suit the glass sheets to be tested. It will be appreciated that glass sheets are manufactured in a vast number of different sizes and shapes. It is therefore necessary that the mechanical frame has a degree of modularity for ease of adaptation to different situations.
  • the frame 5 is clamped to the window 2 using vacuum activated suction pads 5a.
  • the detector head 3 is movable in the X-axis on a transverse rail 6. Movement of the detector head 3 is effected by a stepper motor 7 and belt drive 8.
  • the belt 8 is a toothed belt to substantially eliminate slippage.
  • the X position of the detector head 3 relative to a starting position is therefore determinable by counting of the stepper motor pulses.
  • the transverse rail 6 and detector head 3 are movable in the Y- axis on lateral rails 9.
  • a constant torque motor 10 drives a belt 11 to move the transverse rail 6.
  • An encoder on the belt 11 provides positional information that is combined with the stepper motor information to calculate the coordinates of the detector head 3 at any position within the range of movement of the detector head 3 on the transverse rail 6 and the transverse rail 6 on the lateral rails 9.
  • Support rails 12 provide rigidity to the frame 5.
  • a first embodiment of the detector head 3 consists of at least one detection unit 4 which includes a laser module 13 that emits a beam of coherent radiation.
  • a suitable laser is a semiconductor laser operating at 635nm with a power of approximately 20mW.
  • the laser is modulated at a known frequency, for example 455kHz. It will be appreciated that the specific laser is not critical to the apparatus but will be chosen to suit the particular application.
  • a lens 14 shapes the output of the laser 13 to produce a line of radiation that has a length of approximately 10mm at the surface of the glass and a width of 100 ⁇ m.
  • a prism 15 directs the laser beam 16 into the glass plate at an angle of between 30 and 60 degrees, depending on the thickness of the glass.
  • the detection unit 4 also includes at least one sensor 17 and sensor electronics 18 that measures the radiation scattered from any inclusions in the glass.
  • Silicon photodiodes have been found to be suitable sensors although other arrangements, such as fibreoptic collectors coupled to a photomultiplier tube, CCD cameras, video, and sensor arrays, would also be suitable.
  • the scattered radiation 19 is collected by objective lens 20 and focussing lens 21.
  • the lenses 20, 21 image the scattered radiation onto the sensor 17.
  • a cylindrical lens (not shown) may optionally be used to focus the scattered radiation to a spot at the sensor.
  • the inclusions within the glass act essentially as point source scatterers so the imaging optics can image the scatterer onto the sensor with good efficiency.
  • the optical and physical arrangement spatially filters the radiation reflected or scattered from the front and back surface.
  • the signal obtained from the sensor electronics 18 is primarily due to radiation scattered from within the body of the glass.
  • each inclusion will produce two signals.
  • One signal is produced when the inclusion is illuminated directly by the laser beam 16.
  • Another signal is produced when the laser beam is reflected from the back surface 25 and the reflected beam illuminates the inclusion.
  • the presence of both signals is confirmation that the inclusion occurs in the bulk of the glass rather than on the front or back surface. If only a single signal is detected the position is not recorded.
  • the two signals must be separated by an expected amount (time or distance) for the position of the scatterer to be recorded. This criteria helps to avoid false results caused by anomalous signals generated by strong scatterers on the surface of the glass sheet. A method of determining the size and depth of the inclusion is described below.
  • various filters are employed to improve the signal to noise ratio of the detector head including notch filters to reduce the background.
  • the prism 15 is positioned closely adjacent the surface of the glass.
  • the prism 15 is spring loaded within the detector unit 4 and may move between outer stop 26 and inner stop 27 on bearing 28. This arrangement ensures that the prism 15 is always in a known position relative to the body of the detector unit 4 within the detector head 3. As described below, the detector head 3 rolls on the front surface 24 of the glass 2.
  • the position of the laser beam 16 relative to the glass 2 is therefore always known so the coordinates of an inclusion that causes scattering from within the glass can be accurately recorded.
  • the laser is modulated, preferably at a frequency of 455kHz. Modulating the laser allows for lock-in detection of the scattered radiation and reduces the effect of background light. Furthermore, the rate of scanning may not be constant over the entire scan area and therefore an independent timing mark is required for determining size and depth. This aspect is described in detail below with reference to FIG 10. An alternative embodiment of a detector unit 30 is shown in FIG.
  • the primary difference from the first embodiment is the incorporation of two laser housings 31 , 31 a.
  • Each housing mounts a laser 32, 32a and corresponding focussing lenses 33, 33a.
  • the laser beams 34, 34a are directed to the glass sheet 2 by adjustable mirrors 35, 35a.
  • the mirrors 35, 35a are adjustable to account for different glass thickness.
  • the optimal angle for interrogating the glass sheet with the laser beams is dependent upon the thickness of the glass and the mechanical arrangement of the apparatus, particularly the amount of space available between the laser and the surface of the glass. The inventor has found that an angle of between 40 and 45 degrees is best for 10mm thick glass.
  • a sensor 17 and sensor electronics 18 detect radiation scattered from the bulk of the glass sheet in the manner desc ⁇ bed previously.
  • Lenses 36, 37 collect scattered radiation 38 from the bulk of the glass and direct it to the sensor 17. As with the first embodiment, the front and back surface reflections are not collected by the lenses
  • the optical elements are fixed within the detection unit 30 (except for the adjustability of the mirrors 35, 35a) and the unit 30 is spring loaded to remain close to the surface of the glass. This simplifies the detection unit arrangement compared to the first embodiment.
  • the laser beams 34, 34a are counter- propagating. As the detector head 3 is scanned across the surface of the glass sheet 2, any inclusion will produce scattering from first one laser beam 34 and then the other 34a.
  • the scattering signals are analogous to the forward scattering and backward scattering signals described with reference to FIG 2, but both signals are of approximately the same magnitude. In the embodiment of FIG 2 the backward scattering signal is considerably weaker than the forward scattering signal. For this reason the second embodiment is preferred.
  • the position of a scatterer is only recorded if two signals are detected with an expected separation.
  • the expected separation may be within a certain number of steps of the stepper motor or within a certain period of time, depending on the rate of scan of the detector head.
  • the electronics 18 can then be configured to discriminate between signals from each laser thereby providing an extra dimension of discrimination for accurately recording the position and nature of a detected inclusion.
  • the first embodiment will not be able to correctly identify two inclusions that exist in the glass at a separation equal to the spacing between the laser beams at the inclusion. This is because the second scattering signal could be backwards scattering from the first inclusion or forwards scattering from the second inclusion.
  • frequency discrimination allows the system to differentiate between the received signals.
  • FIG 3 shows the laser beams 34, 34a as counter-propagating and meeting at the rear surface of the glass. It will be appreciated that this is not an essential arrangement.
  • Each laser beam may be directed to the glass at a different angle but recording of the calculation of the position of the detected inclusion will require adjustment to account for the geometric arrangement.
  • the laser beams 34, 34a may be directed at the glass sheet at the same angle but may be arranged to be slightly separated at the rear surface. This will require a slightly wider field of view of the collection optics.
  • FIG 2 and FIG3 show the collection optics as collecting scattered radiation from a distinct point, it will be appreciated that this is merely indicative, in fact the collection optics collect any scattered radiation from within the field of view of the optics.
  • the inventor envisages that both lasers could be adjacent and directing coherent radiation at the glass. This is not a preferred arangement.
  • the detector head 3 may include at least one classification unit 40, shown in FIG 4.
  • the classification unit 40 consists of a CW laser 41 with beam shaping optic 42 that directs a beam 43 of coherent laser radiation towards the known position of an inclusion.
  • the scattered radiation 44 is collected by lens 45 and viewed by a video camera 46.
  • the inventor has found that the scattering from certain inclusions will cause a visible pattern.
  • Inclusions that are smooth scatterers, such as air bubbles, produce specular reflections in all directions. Constructive interference is visible as bright lines in the video image and is indicative of an air bubble or similar specular reflector.
  • a sketch of a typical image of a smooth scatterer is shown in FIG 5.
  • a rough scatterer does not produce a regular constructive interference pattern but instead produces the pattern shown in FIG 6. It can be seen in FIG 6 that there are bright spots that result from constructive interference and dark spots resulting from destructive interference but does not show the regular bands indicative of a specular reflector.
  • the video image may therefore be used to classify a detected inclusion into either a smooth scatterer or a rough scatterer. Since it is known that NiS in glass has a rough texture, any detected inclusion that produces a regular interference pattern can be rejected as not being NiS.
  • the camera may be appropriate to use the camera to capture a wider field of view of the inclusion. This will result in imaging of the inclusion in the manner shown in FIG 7.
  • the information content is essentially the same, but is presented differently.
  • a spectroscopic technique involves spectroscopic analysis of the scattered radiation to categorise the inclusion according to spectroscopic signature.
  • the scattered radiation may have a distinctive Raman scattered signal that categorises the inclusion as NiS.
  • a different categorisation head may be employed that includes a laser emitting radiation in the blue region of the spectrum and a spectrometer that measures fluorescence from NiS.
  • FIG 8 A front view of a typical detector head is shown in FIG 8.
  • the detector head 3 includes three detection units 4 and two classification units 40. Also visible in FIG 8 are rollers 41 that roll on the surface of the glass so that the position of the detector head relative to the glass stays constant, irrespective of imperfections in the glass surface.
  • the detector head 39 for the second embodiment is shown in FIG 9 and is slightly different than the detector head 3 to account for the second laser in the detection unit 30.
  • the detector head includes three detection units and two classification units.
  • Four rollers 42 support the detection head 39 against the glass.
  • the detector head 3 is advanced in the X direction to sweep out a 10 mm strip of the glass for each detection unit, than moved along the Y axis before scanning the next strip. It has been found that scanning rates of 2.5 minutes per square meter is achievable with the preferred embodiment.
  • the coordinates of detected inclusions in the glass are recorded. Not all inclusions will be NiS. In fact, only a small percentage of detected inclusions will be NiS in a typical glass sheet.
  • the detector head is returned to the coordinates of all detected inclusions and the classification unit is used to classify the inclusion.
  • the framework can be fixed with the glass sheet being passed under the framework on a conveyor.
  • the data obtained can be integrated with the glass cutting operation to optimise the cut to eliminate known and suspected NiS inclusions.
  • the operation of the apparatus is similar in either application.
  • the apparatus When testing glass already installed on a building the need for accurate registration requires an additional dimension.
  • the apparatus is fitted to existing equipment for facade maintenance from the outside.
  • the position of the facade maintenance equipment can then be recorded by window number or some other suitable parameter.
  • the time base of the laser pulses to estimate the size and depth of an inclusion from the signal obtained from the scattered radiation.
  • two signals as shown in FIG 10, will be detected for each inclusion for either the first, single laser embodiment, or the second, double laser embodiment.
  • a primary scattering signal 50 will be detected as the first laser beam crosses the inclusion.
  • the second laser beam or the reflection of the laser beam from the back surface will be scattered by the inclusion thereby giving a secondary scattering signal 51.
  • the time difference ⁇ t between the primary scattering signal 50 and the secondary scattering signal 51 indicates the depth of the inclusion.
  • the time that secondary scattering signal 51 occurs after the primary scattering signal 50 depends upon the rate of scan of the detector head 3, 39.
  • each scattered signal has a width ⁇ w.
  • the width of the signal or more specifically the number of scattered laser pulses, gives an indication of the size of an inclusion.
  • a large inclusion will scatter more laser pulses for a given scan rate than a small inclusion.
  • the pulse rate is known, and the rate of advance of the carriage is known, it is a simple matter to calculate the size of the inclusion that equates to a given number of scattered pulses. It has been found that inclusions of greater than or equal to
  • the personal computer provides the signals to stop and start the motors, turn the lasers on and off, and record the coordinates of detected inclusions. These functions could also be performed by a purpose built controller.

Abstract

Light (34, 34a) from one or more lasers (32, 32a) is scanned across a sheet of glass (2). Light is directly incident on inclusions in the glass (2), and indirectly via reflection from the back surface of the glass. Light (38) scattered from an inclusion thus provides at least two scattering signals to a sensor (18), which may be analysed to determine the location, size and depth of the inclusion. Detected inclusions may be revisited and classified, into smooth scatterers such as bubbles and rough scatterers such as material defects, via a camera which records the pattern of scattered laser radiation. Nickel sulphide inclusions can be further classified via spectroscopic analysis. Modulation of the laser beams (34, 34a) improves accuracy and provides a timing mark for the scattered light.

Description

DETECTION OF INCLUSIONS IN GLASS
This invention relates to a method and apparatus for detecting inclusions in glass plate, particularly nickel sulphide (NiS) inclusions. The invention further relates to an apparatus and method for classifying the detected inclusions.
BACKGROUND TO THE INVENTION
The problems associated with inclusions in glass have been known for many years. In particular, it has been known for the last thirty years that heat-strengthened or fully toughened glass plates that have NiS inclusions are subject to spontaneous failure. In short, the presence of NiS inclusions can result in shattering of the glass plate. Unfortunately, these inclusions have proven to be extremely difficult to exclude from the manufacturing process. They are also very difficult to detect. This has led to potentially unsafe glass being used in buildings. The detection and classification of NiS inclusions has proven to be a very difficult problem. A heat-soaking process has been developed in which the manufactured glass is maintained at an elevated temperature. The theory is that almost all glass which contains inclusions that could cause spontaneous failure will fail during the heat soak process. This has proven to be incorrect and there are today many buildings that have been erected using glass that is subject to spontaneous failure due to the effect of NiS inclusions. In fact, a number of instances have been recorded of glass falling from multistory buildings as the result of failure due to NiS inclusions that have not been detected by the heat soak process.
A number of approaches have been proposed for identifying NiS inclusions in glass. One such proposal is found in United States patent number 4697082, assigned to Flachglas Aktiengesellschaft. The Flachglas patent describes a process for testing glass for material defects by illuminating the glass with a laser-produced flying light spot. The forward and backward scattering produced by inclusions in the glass are detected and analysed. Detecting forward and backscattering is not generally practical once the glass has been erected into a building. A single sided approach, normally the outside, is required. Another approach is described in United States patent number 5459330, assigned to Thomson-CSF. This patent describes an apparatus that illuminates successive cross section planes in a sheet of glass and uses a camera to detect reflected radiation. Reflections from the front and back surface of the glass produce two lines in the image that define the boundaries within which an inclusion may be located. Luminous points situated between the two lines are detected as inclusions within the glass. However, the apparatus does not identify the nature of the inclusion nor is the apparatus suited to scanning of glass sheets in situ on a building.
A further apparatus for detecting flaws in glass is described in United States patent number 4597665, assigned to Tencor Instruments. As in the Flachglas device, the Tencor apparatus measures reflected laser light above and below the plane of the glass sheet. In most applications it is impractical to position detectors on both sides of the glass being tested.
OBJECT OF THE INVENTION
It is an object of the invention to provide a method and apparatus for detecting inclusions in glass.
It is a further object of the invention to provide a method and apparatus for classifying inclusions detected in glass sheets.
Further objects will be evident from the following description. SUMMARY OF THE INVENTION
In one form, although it need not be the only or indeed the broadest form, the invention resides in an apparatus for detecting inclusions in glass comprising: one or more lasers emitting one or more coherent beams of radiation directed at the glass at one or more known angles; means for detecting radiation scattered by inclusions within the glass; means for scanning the coherent beams across the glass; and means for recording the location of a scatteπng inclusion. The apparatus may further comprise means for classifying the inclusions. The means for classifying the inclusion suitably comprises a camera that records a pattern of laser radiation scattered from the detected inclusion. The means for classifying may also include means for discriminating a pattern that indicates an inclusion is not NiS The means for classifying may further comprise categorising means for positively categorising the detected inclusions that are NiS. The categorising means is suitably a spectroscopic means that analyses radiation scattered from the inclusion and categorises the scattered radiation by a spectroscopic signature. In a further form, the invention resides in a method of detecting inclusions in glass including the steps of: directing laser radiation at the surface of the glass; scanning the laser radiation over the glass; detecting radiation scattered from inclusions in the glass; and recording the coordinates of the inclusions.
Suitably, the method further includes the steps of detecting first detected radiation scattered from a first beam of laser radiation, detecting second detected radiation scattered from a second beam of laser radiation, and only recording the coordinates of an inclusion if the first detected radiation and the second detected radiation are separated in a known way. The second beam of laser radiation may be a reflection of the first beam.
The method preferably includes the further steps of: viewing the detected inclusion with a camera; and classifying the inclusion according to the pattern of laser radiation scattered from the inclusion.
In preference, the step of classifying classifies the inclusions that are not NiS.
The method may also include the further steps of: spectroscopically analysing the radiation scattered from the detected inclusions; and categorising the identified inclusions as NiS by a spectroscopic signature.
BRIEF DESCRIPTION OF THE DRAWINGS Preferred embodiments of the invention are described with reference to the following figures in which:
FIG 1 is a sketch of an apparatus for detecting inclusions in glass;
FIG 2 is a schematic side view of a first embodiment of a detection unit;
FIG 3 is a schematic side view of a second embodiment of a detection unit;
FIG 4 is a schematic side view of a classification unit;
FIG 5 is a sketch showing the pattern of radiation scattered from a smooth inclusion;
FIG 6 is a sketch showing the pattern of radiation scattered from a rough inclusion;
FIG 7 is a sketch showing another view of the pattern of radiation scattered from a smooth inclusion; FIG 8 is a sketch of a detector head incorporating the detection units of FIG 2; and FIG 9 is a sketch of a detector head incorporating the detection units of FIG 3;
FIG 10 is a timing diagram showing the method of determining size and depth of an inclusion.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT
Referring to FIG 1 there is shown a sketch of an apparatus 1 for detecting inclusions in glass sheets 2. The apparatus 1 includes a detector head 3 that can be scanned over the area of the glass sheet 2. In the preferred embodiment the detector head 3 incorporates multiple detection units, such as 4. The glass sheet 2 may be scanned in situ on a building or on a production line.
The mechanical frame 5 of the apparatus is made from extruded aluminium for strength and light weight. The frame is made to a suitable size to suit the glass sheets to be tested. It will be appreciated that glass sheets are manufactured in a vast number of different sizes and shapes. It is therefore necessary that the mechanical frame has a degree of modularity for ease of adaptation to different situations. The frame 5 is clamped to the window 2 using vacuum activated suction pads 5a.
The detector head 3 is movable in the X-axis on a transverse rail 6. Movement of the detector head 3 is effected by a stepper motor 7 and belt drive 8. The belt 8 is a toothed belt to substantially eliminate slippage. The X position of the detector head 3 relative to a starting position is therefore determinable by counting of the stepper motor pulses.
The transverse rail 6 and detector head 3 are movable in the Y- axis on lateral rails 9. A constant torque motor 10 drives a belt 11 to move the transverse rail 6. An encoder on the belt 11 provides positional information that is combined with the stepper motor information to calculate the coordinates of the detector head 3 at any position within the range of movement of the detector head 3 on the transverse rail 6 and the transverse rail 6 on the lateral rails 9. Support rails 12 provide rigidity to the frame 5.
Turning now to FIG 2, a first embodiment of the detector head 3 consists of at least one detection unit 4 which includes a laser module 13 that emits a beam of coherent radiation. A suitable laser is a semiconductor laser operating at 635nm with a power of approximately 20mW. For reasons described below, the laser is modulated at a known frequency, for example 455kHz. It will be appreciated that the specific laser is not critical to the apparatus but will be chosen to suit the particular application.
A lens 14 shapes the output of the laser 13 to produce a line of radiation that has a length of approximately 10mm at the surface of the glass and a width of 100μm. A prism 15 directs the laser beam 16 into the glass plate at an angle of between 30 and 60 degrees, depending on the thickness of the glass.
The detection unit 4 also includes at least one sensor 17 and sensor electronics 18 that measures the radiation scattered from any inclusions in the glass. Silicon photodiodes have been found to be suitable sensors although other arrangements, such as fibreoptic collectors coupled to a photomultiplier tube, CCD cameras, video, and sensor arrays, would also be suitable.
The scattered radiation 19 is collected by objective lens 20 and focussing lens 21. The lenses 20, 21 image the scattered radiation onto the sensor 17. As the incident laser beam is a line, a cylindrical lens (not shown) may optionally be used to focus the scattered radiation to a spot at the sensor. The inclusions within the glass act essentially as point source scatterers so the imaging optics can image the scatterer onto the sensor with good efficiency. As can be seen in FIG 2, the reflection 22 from the front surface
24 and the reflection 23 from the back surface 25 of the glass 2 are not collected by the lenses 20, 21. Furthermore, scattering from the front surface 24 and back surface 25 is not imaged by the collection optics onto the sensor 17. Although some surface scatter will be collected, the optical and physical arrangement spatially filters the radiation reflected or scattered from the front and back surface. Thus the signal obtained from the sensor electronics 18 is primarily due to radiation scattered from within the body of the glass.
As described in more detail below, each inclusion will produce two signals. One signal is produced when the inclusion is illuminated directly by the laser beam 16. Another signal is produced when the laser beam is reflected from the back surface 25 and the reflected beam illuminates the inclusion. The presence of both signals is confirmation that the inclusion occurs in the bulk of the glass rather than on the front or back surface. If only a single signal is detected the position is not recorded. Furthermore, the two signals must be separated by an expected amount (time or distance) for the position of the scatterer to be recorded. This criteria helps to avoid false results caused by anomalous signals generated by strong scatterers on the surface of the glass sheet. A method of determining the size and depth of the inclusion is described below.
In addition to the geometric arrangement, various filters are employed to improve the signal to noise ratio of the detector head including notch filters to reduce the background. The prism 15 is positioned closely adjacent the surface of the glass. The prism 15 is spring loaded within the detector unit 4 and may move between outer stop 26 and inner stop 27 on bearing 28. This arrangement ensures that the prism 15 is always in a known position relative to the body of the detector unit 4 within the detector head 3. As described below, the detector head 3 rolls on the front surface 24 of the glass 2. The position of the laser beam 16 relative to the glass 2 is therefore always known so the coordinates of an inclusion that causes scattering from within the glass can be accurately recorded.
As mentioned above, the laser is modulated, preferably at a frequency of 455kHz. Modulating the laser allows for lock-in detection of the scattered radiation and reduces the effect of background light. Furthermore, the rate of scanning may not be constant over the entire scan area and therefore an independent timing mark is required for determining size and depth. This aspect is described in detail below with reference to FIG 10. An alternative embodiment of a detector unit 30 is shown in FIG
3. The primary difference from the first embodiment is the incorporation of two laser housings 31 , 31 a. Each housing mounts a laser 32, 32a and corresponding focussing lenses 33, 33a. The laser beams 34, 34a are directed to the glass sheet 2 by adjustable mirrors 35, 35a. The mirrors 35, 35a are adjustable to account for different glass thickness. The optimal angle for interrogating the glass sheet with the laser beams is dependent upon the thickness of the glass and the mechanical arrangement of the apparatus, particularly the amount of space available between the laser and the surface of the glass. The inventor has found that an angle of between 40 and 45 degrees is best for 10mm thick glass.
A sensor 17 and sensor electronics 18 detect radiation scattered from the bulk of the glass sheet in the manner descπbed previously. Lenses 36, 37 collect scattered radiation 38 from the bulk of the glass and direct it to the sensor 17. As with the first embodiment, the front and back surface reflections are not collected by the lenses
Unlike the first embodiment, the optical elements are fixed within the detection unit 30 (except for the adjustability of the mirrors 35, 35a) and the unit 30 is spring loaded to remain close to the surface of the glass. This simplifies the detection unit arrangement compared to the first embodiment. In the second embodiment, the laser beams 34, 34a are counter- propagating. As the detector head 3 is scanned across the surface of the glass sheet 2, any inclusion will produce scattering from first one laser beam 34 and then the other 34a. The scattering signals are analogous to the forward scattering and backward scattering signals described with reference to FIG 2, but both signals are of approximately the same magnitude. In the embodiment of FIG 2 the backward scattering signal is considerably weaker than the forward scattering signal. For this reason the second embodiment is preferred. As with the first embodiment, the position of a scatterer is only recorded if two signals are detected with an expected separation. The expected separation may be within a certain number of steps of the stepper motor or within a certain period of time, depending on the rate of scan of the detector head. In the second embodiment it may be convenient in some applications to modulate each laser beam at a different frequency, say 455kHz for laser beam 34 and 370kHz for laser beam 34a. The electronics 18 can then be configured to discriminate between signals from each laser thereby providing an extra dimension of discrimination for accurately recording the position and nature of a detected inclusion. For example, the first embodiment will not be able to correctly identify two inclusions that exist in the glass at a separation equal to the spacing between the laser beams at the inclusion. This is because the second scattering signal could be backwards scattering from the first inclusion or forwards scattering from the second inclusion. However, frequency discrimination allows the system to differentiate between the received signals.
The schematic diagram of FIG 3 shows the laser beams 34, 34a as counter-propagating and meeting at the rear surface of the glass. It will be appreciated that this is not an essential arrangement. Each laser beam may be directed to the glass at a different angle but recording of the calculation of the position of the detected inclusion will require adjustment to account for the geometric arrangement.
Similarly, the laser beams 34, 34a may be directed at the glass sheet at the same angle but may be arranged to be slightly separated at the rear surface. This will require a slightly wider field of view of the collection optics. Although FIG 2 and FIG3 show the collection optics as collecting scattered radiation from a distinct point, it will be appreciated that this is merely indicative, in fact the collection optics collect any scattered radiation from within the field of view of the optics. The inventor envisages that both lasers could be adjacent and directing coherent radiation at the glass. This is not a preferred arangement.
The detector head 3 may include at least one classification unit 40, shown in FIG 4. The classification unit 40 consists of a CW laser 41 with beam shaping optic 42 that directs a beam 43 of coherent laser radiation towards the known position of an inclusion. The scattered radiation 44 is collected by lens 45 and viewed by a video camera 46. The inventor has found that the scattering from certain inclusions will cause a visible pattern. Inclusions that are smooth scatterers, such as air bubbles, produce specular reflections in all directions. Constructive interference is visible as bright lines in the video image and is indicative of an air bubble or similar specular reflector. A sketch of a typical image of a smooth scatterer is shown in FIG 5.
In contrast, a rough scatterer does not produce a regular constructive interference pattern but instead produces the pattern shown in FIG 6. It can be seen in FIG 6 that there are bright spots that result from constructive interference and dark spots resulting from destructive interference but does not show the regular bands indicative of a specular reflector. The video image may therefore be used to classify a detected inclusion into either a smooth scatterer or a rough scatterer. Since it is known that NiS in glass has a rough texture, any detected inclusion that produces a regular interference pattern can be rejected as not being NiS.
In some instances it may be appropriate to use the camera to capture a wider field of view of the inclusion. This will result in imaging of the inclusion in the manner shown in FIG 7. The information content is essentially the same, but is presented differently.
To further classify the detected inclusions a spectroscopic technique may be employed. The spectroscopic technique involves spectroscopic analysis of the scattered radiation to categorise the inclusion according to spectroscopic signature. For example, the scattered radiation may have a distinctive Raman scattered signal that categorises the inclusion as NiS. Alternatively, a different categorisation head may be employed that includes a laser emitting radiation in the blue region of the spectrum and a spectrometer that measures fluorescence from NiS.
To increase the efficiency of scanning a glass sheet, a number of detection units and classification units can be mounted in a single detector head. A front view of a typical detector head is shown in FIG 8. The detector head 3 includes three detection units 4 and two classification units 40. Also visible in FIG 8 are rollers 41 that roll on the surface of the glass so that the position of the detector head relative to the glass stays constant, irrespective of imperfections in the glass surface.
The detector head 39 for the second embodiment is shown in FIG 9 and is slightly different than the detector head 3 to account for the second laser in the detection unit 30. The detector head includes three detection units and two classification units. Four rollers 42 support the detection head 39 against the glass.
In operation, the detector head 3 is advanced in the X direction to sweep out a 10 mm strip of the glass for each detection unit, than moved along the Y axis before scanning the next strip. It has been found that scanning rates of 2.5 minutes per square meter is achievable with the preferred embodiment.
As mentioned above, the coordinates of detected inclusions in the glass are recorded. Not all inclusions will be NiS. In fact, only a small percentage of detected inclusions will be NiS in a typical glass sheet. After the glass has been scanned the detector head is returned to the coordinates of all detected inclusions and the classification unit is used to classify the inclusion.
In a production line situation the framework can be fixed with the glass sheet being passed under the framework on a conveyor. The data obtained can be integrated with the glass cutting operation to optimise the cut to eliminate known and suspected NiS inclusions. The operation of the apparatus is similar in either application.
When testing glass already installed on a building the need for accurate registration requires an additional dimension. Typically, the apparatus is fitted to existing equipment for facade maintenance from the outside. The position of the facade maintenance equipment can then be recorded by window number or some other suitable parameter. Because the laser radiation is modulated, it is possible to exploit the time base of the laser pulses to estimate the size and depth of an inclusion from the signal obtained from the scattered radiation. It will be appreciated that two signals, as shown in FIG 10, will be detected for each inclusion for either the first, single laser embodiment, or the second, double laser embodiment. A primary scattering signal 50 will be detected as the first laser beam crosses the inclusion. As the scan continues the second laser beam or the reflection of the laser beam from the back surface will be scattered by the inclusion thereby giving a secondary scattering signal 51. The time difference Δt between the primary scattering signal 50 and the secondary scattering signal 51 indicates the depth of the inclusion. The time that secondary scattering signal 51 occurs after the primary scattering signal 50 depends upon the rate of scan of the detector head 3, 39.
As shown in FIG 10, each scattered signal has a width Δw. The width of the signal, or more specifically the number of scattered laser pulses, gives an indication of the size of an inclusion. A large inclusion will scatter more laser pulses for a given scan rate than a small inclusion. Given that the pulse rate is known, and the rate of advance of the carriage is known, it is a simple matter to calculate the size of the inclusion that equates to a given number of scattered pulses. It has been found that inclusions of greater than or equal to
50μm size can be detected with the apparatus. Furthermore, these inclusions can be classified as not NiS or otherwise with a high degree of accuracy.
It is convenient to operate the apparatus from a conventional personal computer. The personal computer provides the signals to stop and start the motors, turn the lasers on and off, and record the coordinates of detected inclusions. These functions could also be performed by a purpose built controller.
Throughout the specification the aim has been to describe the invention without limiting the invention to any specific combination of features.

Claims

1. An apparatus for detecting inclusions in glass comprising: one or more lasers emitting one or more coherent beams of radiation directed at the glass at one or more known angles; means for detecting radiation scattered by inclusions within the glass; means for scanning said coherent beams across the glass; and means for recording the location of a scattering inclusion.
2. The apparatus of claim 1 comprising two lasers, said two lasers being positioned either side of said means for detecting radiation scattered by inclusions, and said two lasers directing coherent beams of radiation at said glass so as to be counter-propagating.
3. The apparatus of claim 1 further comprising means for classifying a detected inclusion.
4. The apparatus of claim 1 further comprising means for classifying a detected inclusion wherein said means for classifying comprises a camera that records a pattern of laser radiation scattered from the detected inclusion.
5. The apparatus of claim 1 further comprising means for classifying a detected inclusion wherein said means for classifying the detected inclusion comprises a camera that records a pattern of laser radiation scattered from the detected inclusion and means for discriminating a pattern that indicates an inclusion is not NiS.
6. The apparatus of claim 1 further comprising means for classifying a detected inclusion, said means for classifying comprising a continuous wave laser emitting a coherent beam of radiation at the detected inclusion and a camera that records a pattern of laser radiation scattered from the detected inclusion.
7. The apparatus of claim 1 further comprising categorising means for categorising the detected inclusions that are NiS.
8. The apparatus of claim 7 wherein the categorising means is a spectroscopic means that analyses radiation scattered from the inclusion and categorises the scattered radiation by a spectroscopic signature.
9. The apparatus of claim 1 where said one or more lasers are modulated at a known frequency.
10. The apparatus of claim 1 wherein there are two lasers each modulated at a known but different frequency.
11. A method of detecting inclusions in glass including the steps of: directing laser radiation at the surface of the glass; scanning the radiation over the glass; detecting radiation scattered from inclusions in the glass; and recording the coordinates of the detected inclusions.
12. The method of claim 11 wherein the step of detecting radiation scattered from inclusions includes the steps of: detecting first detected radiation scattered from a first beam of laser radiation; detecting second detected radiation scattered from a second beam of laser radiation; and only recording the coordinates of an inclusion if the first detected radiation and the second detected radiation are separated in a known way.
13. The method of claim 12 wherein said second beam is a reflection of said first beam.
14. The method of claim 11 further including the steps of : viewing the detected inclusion with a camera; and classifying the inclusion according to the pattern of laser radiation scattered from the detected inclusion.
15. The method of claim 11 further including the steps of: viewing the detected inclusion with a camera; classifying the inclusion according to the pattern of laser radiation scattered from the detected inclusion; and identifying the inclusions that are not NiS.
16. The method of claim 11 further including the steps of : spectroscopically analysing the radiation scattered from the detected inclusions; and categorising the identified inclusions as NiS by a spectroscopic signature.
EP00958046A 1999-09-02 2000-09-01 Detection of inclusions in glass Withdrawn EP1210586A4 (en)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
AUPQ2622A AUPQ262299A0 (en) 1999-09-02 1999-09-02 Detection of inclusions in glass
AUPQ992622 1999-09-02
PCT/AU2000/001032 WO2001018532A1 (en) 1999-09-02 2000-09-01 Detection of inclusions in glass

Publications (2)

Publication Number Publication Date
EP1210586A1 true EP1210586A1 (en) 2002-06-05
EP1210586A4 EP1210586A4 (en) 2002-12-04

Family

ID=3816792

Family Applications (1)

Application Number Title Priority Date Filing Date
EP00958046A Withdrawn EP1210586A4 (en) 1999-09-02 2000-09-01 Detection of inclusions in glass

Country Status (7)

Country Link
EP (1) EP1210586A4 (en)
JP (1) JP2003508786A (en)
CN (1) CN1377462A (en)
AU (1) AUPQ262299A0 (en)
CA (1) CA2384005A1 (en)
HK (1) HK1049042A1 (en)
WO (1) WO2001018532A1 (en)

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7142295B2 (en) * 2003-03-05 2006-11-28 Corning Incorporated Inspection of transparent substrates for defects
DE102004005019A1 (en) * 2004-01-30 2005-08-18 Isra Glass Vision Gmbh Method for determining the depth of a defect in a glass ribbon
EP1794577A4 (en) * 2004-09-17 2010-10-06 Wdi Wise Device Inc Optical inspection of flat media using direct image technology
GB0606217D0 (en) * 2006-03-29 2006-05-10 Pilkington Plc Glazing inspection
WO2009031420A1 (en) * 2007-09-04 2009-03-12 Asahi Glass Company, Limited Method and device for detecting micro foreign matter within transparent plate
DE102013002602B4 (en) 2013-02-15 2022-05-05 Hegla Boraident Gmbh & Co. Kg Method and device for detecting particles in glass
DE102013107215B3 (en) 2013-07-09 2014-10-09 Heraeus Quarzglas Gmbh & Co. Kg Method for producing a mirror substrate blank made of titanium-doped silica glass for EUV lithography, and system for determining the position of defects in a blank
US10928330B2 (en) 2018-03-07 2021-02-23 Guardian Glass, LLC Method and system for detecting inclusions in float glass based on wavelength(s) analysis
US10481097B1 (en) 2018-10-01 2019-11-19 Guardian Glass, LLC Method and system for detecting inclusions in float glass based on spectral reflectance analysis
CN116835863B (en) * 2023-07-14 2023-12-12 连云港双石特种玻璃有限公司 Processing equipment capable of accurately controlling thickness of special glass and application method thereof

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3814946A (en) * 1972-12-04 1974-06-04 Asahi Glass Co Ltd Method of detecting defects in transparent and semitransparent bodies
US4025201A (en) * 1975-04-21 1977-05-24 Ball Brothers Service Corporation Method and apparatus for video inspection of articles of manufacture by decussate paths of light
US4681442A (en) * 1984-05-14 1987-07-21 International Business Machines Corporation Method for surface testing
US5235400A (en) * 1988-10-12 1993-08-10 Hitachi, Ltd. Method of and apparatus for detecting defect on photomask
JPH07294422A (en) * 1994-04-27 1995-11-10 Mitsubishi Materials Corp Detecting method for surface vicinity crystal defect and device therefor

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4136961A (en) * 1977-07-01 1979-01-30 Corning Glass Works Method and apparatus for detection of inclusions in glass article or the like
US4401893A (en) * 1981-07-29 1983-08-30 Intec Corporation Method and apparatus for optically inspecting a moving web of glass
DE3565893D1 (en) * 1984-12-14 1988-12-01 Flachglas Ag Method and device for inspecting transparent strip material, in particular flat glass ribbons
US4914309A (en) * 1987-05-27 1990-04-03 Nippon Sheet Glass Co., Ltd. Discriminating type flaw detector for light-transmitting plate materials
DE3800053A1 (en) * 1988-01-04 1989-07-13 Sick Optik Elektronik Erwin OPTICAL ERROR INSPECTION DEVICE
FR2681429B1 (en) * 1991-09-13 1995-05-24 Thomson Csf METHOD AND DEVICE FOR INSPECTING GLASS.
AT402860B (en) * 1993-06-22 1997-09-25 Oesterr Forsch Seibersdorf METHOD AND DEVICE FOR TESTING TRANSPARENT OBJECTS
NL1006378C2 (en) * 1997-06-23 1998-12-24 Tno Method and device for inspecting an object with respect to disturbances.

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3814946A (en) * 1972-12-04 1974-06-04 Asahi Glass Co Ltd Method of detecting defects in transparent and semitransparent bodies
US4025201A (en) * 1975-04-21 1977-05-24 Ball Brothers Service Corporation Method and apparatus for video inspection of articles of manufacture by decussate paths of light
US4681442A (en) * 1984-05-14 1987-07-21 International Business Machines Corporation Method for surface testing
US5235400A (en) * 1988-10-12 1993-08-10 Hitachi, Ltd. Method of and apparatus for detecting defect on photomask
JPH07294422A (en) * 1994-04-27 1995-11-10 Mitsubishi Materials Corp Detecting method for surface vicinity crystal defect and device therefor

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of WO0118532A1 *

Also Published As

Publication number Publication date
CA2384005A1 (en) 2001-03-15
HK1049042A1 (en) 2003-04-25
JP2003508786A (en) 2003-03-04
CN1377462A (en) 2002-10-30
AUPQ262299A0 (en) 1999-09-23
WO2001018532A1 (en) 2001-03-15
EP1210586A4 (en) 2002-12-04

Similar Documents

Publication Publication Date Title
AU616731B2 (en) Method and apparatus for monitoring the surface profile of a moving workpiece
AU646719B2 (en) Sorting diamonds or other minerals
US5132791A (en) Optical sheet inspection system
US6781705B2 (en) Distance determination
US20070163352A1 (en) Detecting rail defects
US4402607A (en) Automatic detector for microscopic dust on large-area, optically unpolished surfaces
JPH0695075B2 (en) Surface texture detection method
JPH10213539A (en) Detecting device for near-wafer particulate in semiconductor device manufacturing equipment
JPH04331315A (en) Light-detecting system
JP2534453B2 (en) Surface inspection method and its equipment
EP1210586A1 (en) Detection of inclusions in glass
JP2002062267A (en) Device for inspecting defect
JPH0760262B2 (en) Reticle inspection method and device
US7245366B2 (en) Surface inspection method and surface inspection apparatus
US5724140A (en) Method and apparatus for determining the quality of flat glass sheet
AU765183B2 (en) Detection of inclusions in glass
JPH07218419A (en) Light scattering type instrument and method for measuring particles in wide area
CN1367754A (en) Edge detector
CA2282015C (en) Light-scanning device
JPH06281593A (en) Method and apparatus for inspecting surface
WO1999064845A1 (en) Defect detecting unit
TWI485392B (en) Foreign body inspection device and inspection method
JP2003028808A (en) Method and apparatus for surface inspection
KR102208064B1 (en) Improved real-time measurement device for vehicle exhaust gas
SU1296837A1 (en) Device for checking geometric dimensions and flaws of specimens with dissipating surfaces

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

17P Request for examination filed

Effective date: 20020315

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LI LU MC NL PT SE

AX Request for extension of the european patent

Free format text: AL;LT;LV;MK;RO;SI

A4 Supplementary search report drawn up and despatched

Effective date: 20021018

AK Designated contracting states

Kind code of ref document: A4

Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LI LU MC NL PT SE

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE APPLICATION IS DEEMED TO BE WITHDRAWN

18D Application deemed to be withdrawn

Effective date: 20040331

REG Reference to a national code

Ref country code: HK

Ref legal event code: WD

Ref document number: 1049042

Country of ref document: HK