SP05-072
APPARATUS AND METHOD FOR MEASURING A GLASS SHEET
BACKGROUND OF THE INVENTION
FIELD OF THE INVENTION
[0001] This invention is directed to a method of measuring a substantially planar sheet. The invention is particularly useful for measuring warp in thin sheets of glass, such as are used for glass substrates for flat panel display devices.
TECHNICAL BACKGROUND
[0002] Liquid crystal displays (LCDs) are flat panel display devices that include thin, flat glass panes which have pristine, defect-free surfaces. At least several thin panes of glass are sealed together to form an envelope in a display device. It is highly desirable that the glass sheets which comprise these displays do not exhibit surface shape (out-of-plane deformation) so that proper registration is maintained between the glass layers when the display device is assembled. More simply put, it is highly desirable that the glass sheet be flat. Out-of-plane deformation (flatness) is typically referred to as warp.
[0003] Elimination of warp in sheet glass destined for flat panel display applications is an ongoing challenge. A necessary tool in that effort is the ability to accurately measure warp. Numerous methods of measuring warp having varying levels of sophistication, exist in the art. However, few are directed to the measurement of very large, very thin sheets of a brittle, elastic material. One consideration in the development of an effective warp measurement is the manner in which the glass sheet is presented for measurement. That is, how the glass sheet is supported (presented) during the measurement process. Because the glass sheet used for display devices is very thin, on the order of less than about a millimeter, contact with the glass sheet tends to impart a separate source of warp which affects the measurement. This is exacerbated by the fact that demand for very large sheets of display glass is increasing. Today, glass sheets in excess of several square meters in size are being manufactured, and such large, thin sheets must be measured for warp.
[0004] Ideally, large glass sheets would be presented in a contact-free and gravity-free environment. As this is difficult to achieve, particularly in a terrestrial production environment, it is highly desirable that the method of presentation minimizes additional warp of the sheet.
[0005] Prior art off-line glass sheet warp measurement methods have included laying the glass sheet on top of a flat base, such as a marble table top which has been polished smooth and flat. However, the flat base surface is difficult to maintain particulate-free. The presence of dust or other particulate is capable of producing an erroneous measurement of a glass sheet for which warp measurements are expected to have an accuracy to within a few micrometers (microns).
[0006] It would be beneficial to find an apparatus for presenting (supporting) the glass sheet which may advantageously minimize the surface are in contact with the sheet, eliminate or minimize the potential for contamination, and provide sufficient flexibility to accommodate changing measurement needs.
SUMMARY
[0007] Embodiments of the present invention provide a method for measuring the shape (warp) of a substantially planar sheet using support members resting on a base. The glass sheet to be measured is supported by the support members at what is essentially point contact with each support member contacting the glass sheet.
[0008] In one embodiment of the inventive apparatus, the apparatus comprises a base and a plurality of support members positioned on the base, each support member being adapted such that contact between the substantially planar sheet and any one of the plurality of support members is a point contact only. Conventional non-contact measurement means may be employed for measuring an attribute of the sheet, such as, for example, a laser ranging device for measuring a distance from the measurement device to a surface of the sheet. [0009] In accordance with the present embodiment, the pitch between contact points, that is, the distance between the point contact of one repositionable support member with the sheet to be measured and an adjacent point contact, is preferably uniform, and preferably less than about 10 cm; more preferably less than about 5 cm.
[0010] In some embodiments, to maintain a uniform pitch, and ensure that the support members do not move relative to each other during the course of a measurement, it is desirable to employ a support member restraining device. The support member restraining device preferably has a least one opening for receiving the support members and maintaining a positional relationship between the support members. The plurality of support members are preferably arranged in repeating unit cells, and the restraining device preferably comprises a plurality of openings for receiving the support members. It is desirable, but not necessary, that the restraining device be electrically grounded.
[0011] The restraining device may also take the form of an encircling band or the like which conforms the support members according to a particular arrangement, the support members being arranged within the perimeter of the encircling restraining device. Thus, only a subset of the support members, those along the periphery of the array of support members, are in contact with the restraining device.
[0012] In accordance with the present embodiment, the maximum height of each support member above the base deviates from a predetermined value by less than about 10 μm. [0013] A method of measuring warp in a glass sheet is also described comprising the steps of positioning the substantially planar sheet on a plurality of support members, each support member being adapted such that contact between the substantially planar sheet and any one
of the plurality of support members is point contact only, measuring a distance from a sensor to a surface of the substantially planar sheet at a plurality of locations on the sheet and using the distance measurements to determine a warp of the sheet. In some embodiments of the method, it may be desirable to calibrating the movement of the sensor. Calibration comprising the steps of a) positioning a calibration flat in contact with a first unit cell of support members; b) measuring a distance from the sensor to the calibration flat; c) positioning the calibration flat on a second unit cell; and d) repeating steps a)-c). Preferably, the calibration flat is in contact with only a single unit cell at a time.
[0014] The invention will be understood more easily and other objects, characteristics, details and advantages thereof will become more clearly apparent in the course of the following explanatory description, which is given, without in any way implying a limitation, with reference to the attached Figures. It is intended that all such additional systems, methods, features and advantages be included within this description, be within the scope of the present invention, and be protected by the accompanying claims.
BRIEF DESCRIPTION OF THE DRAWINGS
[0015] FIG. 1 is a cross sectional view from the side of an apparatus for presenting a glass sheet for warp measurement.
[0016] FIG. 2 is a top view of a carriage plate for restraining movement of the bearings.
[0017] FIG. 3 is a cross sectional view of the carriage plate of FIG. 2.
[0018] FIG. 4 is a top view of a latticework restraint for restraining the bearings.
[0019] FIG. 5 is a detail view, in cross section as seen from an edge, of a portion of the lattice work restraint of FIG. 4 showing a bearing, a collar and several struts.
[0020] FIG. 6 is a top view of another device for restraining movement of the bearings wherein the bearings are close-packed and restrained within a frame.
[0021] FIG. 7 is a portion of an apparatus for presenting a glass sheet for warp measurement, showing in perspective a bearing placed within a well formed by mounting a washer on the base.
[0022] FIG. 8 is a top view of the apparatus of FIG. 1 including the measurement device mounted on a moving x-y stage.
[0023] FIG. 9 is a side view of an embodiment according to the present invention showing the translation rails and Z-axis stage for movement of the sensor in a plane above and parallel to the glass sheet to be measured, as well as in a direction perpendicular to the glass sheet.
[0024] FIG. 10 is a top view of the carriage plate of FIG. 2 showing the movement of a calibration flat among unit cells of the bearings for determining the APS.
[0025] FIGS. 11a and l ib illustrate a top view of several unit cells, the first unit cell a square consisting of four support members, and the second unit cell a trapezoid, also consisting of four support members.
[0026] FIG. 12a depicts a perspective view of a pyramidal support member.
[0027] FIG. 12b illustrates a footprint of the pyramidal support member of FIG. 12a, and depicts the apex of the pyramidal support member projected vertically downward onto the footprint, the apex representing the location of the point contact between the support member and the glass sheet to be measured.
[0028] FIG. 13 is a perspective view of a portion of a base indicating how a plurality of pyramidal support members of the kind illustrated in FIGS 12a and 12b may be arranged, with the black dots representing the locations of the projected apexes of FIG. 12b.
Detailed Description
[0029] In the following detailed description, for purposes of explanation and not limitation, example embodiments disclosing specific details are set forth to provide a thorough understanding of the present invention. However, it will be apparent to one having ordinary skill in the art, having had the benefit of the present disclosure, that the present invention may be practiced in other embodiments that depart from the specific details disclosed herein. Moreover, descriptions of well-known devices, methods and materials may be omitted so as not to obscure the description of the present invention. Finally, wherever applicable, like reference numerals refer to like elements.
[0030] FIG. 1 shows an embodiment of an apparatus for measuring a substantially planar sheet of material, such as a brittle material, e.g. glass or glass ceramic. Typically, such measurements are directed to determining out-of-plane deviation of the sheet (i.e. flatness, or warp). The following description will be presented in terms of measuring warp of a substantially planar sheet of glass. However, the skilled artisan will realize that the support apparatus described herein may be applicable to other uses, and is not limited to the measurement of warp in a glass sheet.
[0031] The apparatus of FIG. 1 comprises a base 10, and a plurality of spherical members 12 (hereinafter bearings 12). Base 10 is typically made of granite, but may be formed from other dimensionally stable materials, or be constructed in a dimensionally stable manner. For example, an optical tabletop, breadboard or the like, as is used to mount laboratory optical components, may be employed. Such tabletops are readily commercially available. A suitable base for measuring large sheets of glass, e.g. on the order of several square meters or larger, may require custom manufacture. By dimensionally stable what is meant is the base does not exhibit noticeable distortion during the period in which a measurement is made. It is desirable that base 10 be sturdily mounted in a manner which does not impart distortion or vibration to the top surface of the base. For example, base 10 may be mounted in a metal frame and supported by pneumatic legs to dampen or eliminate vibration from being transferred from the surrounding environment, e.g. the ground or the floor, to the glass sheet being measured. Preferably, upper surface 14 of base 10 is flat to within 15 μm. That is, upper surface 14 deviates from an ideal plane by no more than about 15 μm at any point on surface 14. It is also desirable that the base be sufficiently stiff that the base imparts few mechanical resonances to the glass sheet being measured and does not sag, either under the
weight of the base itself or the glass being measured. There should be essentially no measurable sag. A granite base approximately 15 cm thick, for example, has been found to be adequate to eliminate sag.
[0032] Bearings 12 may, for example, be precision ball bearings formed of a suitable metal, such as stainless steel, chromium or other hard metal. The bearings should have a maximum diameter tolerance of 10 μm or less. That is, the diameter of each bearing should have a maximum diameter of d + 5 μm where d is a predetermined nominal diameter. The nominal diameter of each bearing is dependent, inter alia, upon the desired pitch of the bearings, explained in more detail below.
[0033] To ensure that temperature changes do not significantly alter measurement of a glass sheet, it is desirable that the ambient temperature surrounding the inventive apparatus be maintained within a range of ± I0F during the time in which a measurement is made. However, the allowable deviation in temperature is also dictated by the degree of accuracy and precision desired in the measurement. A typical measurement temperature is 68°F ± I0F. [0034] Returning to FIG. 1, bearings 12 are positioned on base surface 14. Preferably, each bearing 12 is in direct contact with surface 14 at a single point - the base-bearing point-of- contact. Each base-bearing point-of-contact is a predetermined distance Λ from its nearest neighboring base-bearing point-of-contact. Distance Λ is termed the pitch. It is preferred that bearings 12 are arranged on base surface 14 in a geometric pattern with a uniform pitch. For example, bearings 12 may be positioned on base surface 14 in a square grid pattern (i.e. the bearings positioned at the four corners of a square). Alternatively, other geometric patterns may be used, such as concentric circles, hexagonal, etc. A typical pitch Λ is less than about 3 cm, but may vary depending upon customer requirements and the thickness of the glass sheet. Generally, the thinner the glass sheet, the smaller the pitch needed to ensure proper support for the glass sheet.
[0035] To prevent movement of the bearings during the course of measuring the glass sheet, a restraining device may be employed. Carriage plate 16, as best illustrated in FIG. 2, is a plate of suitable material having a plurality of openings 18 extending through the thickness of the carriage plate and into which bearings 12 may be inserted such that at least a portion of each bearing extends from and above the carriage plate. In the embodiment shown in FIGS. 1 and 2, one side of carriage plate 16 rests on base 10. Suitable materials for carriage plate 16 include any material that is capable of maintaining each bearing in a predetermined relationship with the other bearings in the array while the bearings are supporting a glass
sheet. For example, carriage plate 16 may be comprised of any one of a variety of different polymers (e.g. Delrin®, manufactured by DuPont). Polymers have the advantage of being light in weight, easy to machine and relatively inexpensive. On the other hand, carriage plate 16 may be formed from a metal, such as aluminum. Advantageously, an electrically conductive restraining device, such as afforded by aluminum or other metals, may be electrically grounded, thereby minimizing static electric buildup that could attract dust to the bearing surfaces and provide erroneous measurement results.
[0036] In the embodiment depicted in FIG. 1 and 3, openings 18 in carriage plate 16 have beveled inside walls 20 to facilitate placement of carriage plate 16 over bearings 12. Advantageously, the beveled side walls 20 aid in preventing contamination of the bearing surfaces by preventing dust from entering the opening. That is, the narrow portion of each opening at the top side of the carriage plate (and as depicted in FIG. 1) fits closely around each bearing. However, such a feature is not necessary, and the inside walls 20 of openings 18 could be cylindrical instead of beveled. That is, the opening on each side of carriage plate 16 could be of equal size, with the walls of the opening perpendicular to each side surface (face) of the carriage plate.
[0037] Other mechanisms for restraining movement of bearings 12 relative to each other may be used. For example, a latticework as illustrated in FIG. 4 may be formed, the latticework comprising connecting members or struts 22 and collars 24. Struts 22 connect the plurality of collars 24 and maintain a predetermined distance between the collars. Bearings 12 are inserted into collars 24, such as by snapping the collars over the bearings. (A significant space is shown between bearings and collars in FIG. 4 for illustrative purposes, i.e. to better show the relationship between bearings and collars.) Collars 24 preferably have an arcuate inner surface 26, wherein the largest inner diameter of the collars is positioned approximately about the circumference of each bearing, and is slightly larger than the circumference. Thus, the collar latticework is suspended above the base surface by the bearings and each bearing 12 is retained within its respective collar 24 but is preferably free to rotate within the collar. A detailed cross sectional view of a collar placed over a bearing, the bearing resting on surface 14 of base 10 is shown in FIG. 5.
[0038] Advantageously, because the support members are not permanently affixed to the base, they may be repositioned into different configuration (e.g. different shape unit cells) simply by employing different restraining devices. For example, carriage plate 16 may have more or fewer openings, with different spacings between openings.
[0039] In another embodiment depicted in FIG. 6, the plurality of bearings 12 may be in a close-packed configuration in that the bearings are in contact with each other in a fashion analogous to racked pool balls. In the configuration shown in FIG. 6, bearings 12 are surrounded by frame 26. Frame 26 maintains the formation of bearings 12 in a predetermined array with a pitch equal to the circumferential diameter of the bearings (assuming each bearing 12 has the same equatorial diameter).
[0040] Other methods of restraining lateral movement of the bearings (i.e. movement parallel to the plane of the base) may also be used and are intended to fall within the scope of the present invention. For example, cylindrical sections (hereinafter "washers" 28), may be secured to the top surface of base 10 in a desired, predetermined pattern, such as by an adhesive, or by welding in the case of metal washers 28. A detailed view in perspective is shown in FIG. 7 indicating a shallow well 30 formed at each predetermined location by the base and the washer. A bearing 12 is placed in each well 30. The depth of each well 30 is such that each bearing is in contact with the base surface at a point, as before, and lateral movement of the bearing is minimized by the presence of the washer and at least a portion of the bearing extends above the washer. Alternatively, the well may have a diameter smaller than a diameter of the bearing such that the bearing rests on top of the washer and the bearing is not in contact with the base, or the diameter of the well may be such that the bearing is in contact with both the washer and the base simultaneously.
[0041] In yet another embodiment, concave dimples (not shown) may be formed in the surface of the base for holding the bearings in position, or the bearings may simply be affixed in position by cementing the bearings to base surface 14 with an adhesive. Other methods of maintaining the position of the bearings are certainly within the capability of one having ordinary skill in the art, and are intended to fall within the scope of the present invention. [0042] In accordance with the present embodiment, the inventive apparatus further comprises a measurement device located over the base and support members. FIG. 8 shows sensor 32 mounted on a Cartesian rail system such that the measurement device may be moved in a plane parallel with the base surface, i.e. in an x-y plane (wherein the x-y plane represents orthogonal distances in a plane parallel with the base surface and the z-direction (FIG. 9) represents a direction orthogonal to the base surface). Generally the x-y directions correspond to the length and width of the glass sheet to be measured. In this context the length and width are arbitrary designations representing perpendicular sides of a rectangular sheet of glass. Sensor 32 may be moved along rails 34 in the x direction by linear stepper motors for example (not shown), and similarly along gantry 36 in the y direction. However,
any suitable translational technique as is known in the art may be used (i.e. which allows the measurement device to translate to predetermined coordinates above glass sheet 38). For example, translation may be based on a different coordinate system, such as polar. Sensor 32 may be any non-contact device suitable for measuring a distance as is known in the art, and may comprise, for example, a conventional laser ranging device, an interferometric device, or an acoustic ranging device. Also included is a translational stage 40 mounted on gantry 36 operating in the z direction (into the paper in the figure) such that the distance between sensor 32, mounted on gantry 36, and the glass sheet to be measured may be varied. [0043] To measure a sheet of glass, an imaginary reference surface (the artificial plane surface - APS) is first identified as the plane which rests upon the top surface of each support member, e.g. the APS rests atop each bearing 12. The tolerance for flatness of the APS is defined by the flatness of base surface 14 and the height tolerance of the support members, e.g. the diameter tolerance of bearing 12. Thereafter, deviation of sensor should be determined along the range of movement of the sensor. A square grid pattern for support members in the form of bearings is next described for illustrative purposes. [0044] A small reference sheet of glass, generally designated as calibration flat 42, having a known flatness is placed in contact with the top of the support members representing the smallest square unit cell - unit cell A1. One skilled in the art will realize that a square grid pattern having a single point contact between each support member and the glass under measurement will encompass many different four-bearing arrangements of various sizes. One of ordinary skill in the art will understand that by "point contact" what is meant is that the support member is in contact with a particular body only over a very small (infinitesimal) surface area, e.g. the point of a pin, the apex of a pyramidal structure, and so forth. In the case of a spherical support member in contact with an essentially planer surface, contact between the surface and the spherical member is a point contact. The calibration flat should be in contact only with the support members which make up the unit cell. A unit cell in the context of the present description represents a square formed from contact with the smallest number of bearings which forms a square - four, as illustrated in FIG. 10. Other unit cell shapes are possible. FIG. l la depicts one four-bearing unit cell A having a square shape, as in FIG. 10, whereas FIG. l ib illustrates a four-bearing unit cell B having a trapezoidal shape. Sensor 32 is translated to a position directly over calibration flat 42 in a first position (designated by 42a in FIG. 10) and the distance from the upper surface of the calibration flat, preferably at the center of the calibration flat, to the measurement device is determined. The calibration flat is then moved to a next unit cell A2 and the measurement is repeated (in this
position the calibration flat is indicated by 42b). Successive placement of the calibration flat on the unit cells of the total bearing grid yields data from which deviation in the vertical, z axis, of sensor 32 over the range of motion of sensor 32 may be determined. That is, a distance from the sensor to a known reference surface (the calibration flat) is made at a plurality of locations covering the range of motion of the sensor in the horizontal x-y plane to determined z-axis deviation of the sensor.
[0045] Once the vertical deviation of the sensor has been determined over a range of movement of the sensor, the calibration flat is removed from the bed of support members and glass sheet 38, the sheet of glass to be measured, is placed in position on top of the bearings 12. The measurement sensor is again used to measure the distance from the sensor to the upper surface of glass sheet 38 at a plurality of discrete points (coordinates) on the glass sheet. In this instance, the glass sheet is maintained stationary while the sensor is moved to each measurement location over the glass sheet. The larger the number of measurement points, the greater the accuracy with which the shape of the sheet (i.e. the deviation of the sheet from a plane surface) can be determined. For example, as many as 10,000 discrete measurements may be made. The discrete measurements are used to calculate a preliminary sheet shape, and the sensor deviation in the z axis is subtracted out to remove any influence in the shape of the sheet which might be contributed by differences in bearing heights. The result is an overall shape of the glass sheet, i.e. the out-of-plane deviation of the sheet as a function of position on the sheet (warp).
[0046] Although the preceding embodiments utilized glass sheet supporting members having a spherical shape, other shapes could be utilized. For example, the supporting members could be pyramidal, with the base of each pyramid in contact with the base, and the apex of each pyramid in contact with the glass sheet (i.e. a point contact) to be measured. As in the case of the bearings, the individual pyramids are arranged in a predetermined geometric pattern on the base. A plurality of such support members could be arranged on base 10. FIG. 12a shows a support member 12 in the shape of a three-sided pyramid, and FIG. 12b illustrates the triangular contact pattern 42 (footprint) of the base of triangular support member 12 on base surface 14. Black dot 44 represents the apex of the pyramid projected onto footprint 42. Placement of support members 12 could be as depicted in FIG. 13, wherein each dot represents the placement of projected apex 44. This placement scheme would also be valid for other shapes, such as a four sided pyramidal shape. [0047] It should be emphasized that the above-described embodiments of the present invention, particularly any "preferred" embodiments, are merely possible examples of
implementations, merely set forth for a clear understanding of the principles of the invention. Many variations and modifications may be made to the above-described embodiments of the invention without departing substantially from the spirit and principles of the invention. All such modifications and variations are intended to be included herein within the scope of this disclosure and the present invention and protected by the following claims.