Optics: measuring and testing – By light interference – For dimensional measurement
Reexamination Certificate
1997-12-01
2001-10-09
Font, Frank G. (Department: 2877)
Optics: measuring and testing
By light interference
For dimensional measurement
C356S519000
Reexamination Certificate
active
06301009
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention relates to surface finishing machines and, more particularly, relates to a method and apparatus for measuring a dimensional characteristic of a workpiece, in-situ, while it is positioned on a surface finishing machine and even while it is being polished or otherwise finished by the surface finishing machine. In a particularly preferred embodiment, variations in the optical thickness of the workpiece are measured by an interferometer, and these measurements are used to control the finishing process to obtain a desired workpiece dimensional characteristic.
2. Discussion of the Related Art
Surface finishing machines are used in numerous applications for polishing, grinding, or otherwise machining one or more surfaces of workpieces made from glass, silicon, metals, ceramics, etc. A surface finishing machine may be adapted to finish a workpiece to a desired thickness, a desired transmitted wavefront quality, a desired smoothness, and/or a desired profile.
One type of surface finishing machine to which the invention is particularly (but not exclusively) applicable is a ring polisher. A ring polisher, sometimes known as a Crane polisher, a lapping machine, a planetary polishing machine, or a continuous polishing machine, is characterized by a large rotating turntable presenting an upper work surface or lap formed by a pitch annulus. Disposed either directly on top of the lap or slightly above the lap are a plurality of work rings which are driven to rotate by rollers. At least one workstation is located in the interior of each work ring for receiving a workpiece. A large, flat conditioning tool is disposed on the surface of the lap and can be rotated and translated radially to adjust the flatness of the lap.
In use, a workpiece is deposited in a corresponding workstation (i.e., within a corresponding work ring), the lap is flooded with a slurry that contains a polishing grit or finishing grit, and the turntable is driven to rotate, thereby causing the work rings to rotate. The workpiece moves with the work ring over the lap at an angular velocity which is synchronous with the velocity of the lap so that the time averaged relative velocity between the lap and every location on the work surface of the workpiece remains constant. The relative movement between the workpiece and the lap polishes the front surfaces of the workpiece via chemo-mechanical abrasion by the grit in the slurry.
The polishing process must be carefully monitored and controlled to achieve the desired results. For instance, if the workpiece is a glass element designed for use as an optical quality element, the topography of the front or work surface must be held to a designated smoothness and profile within sub-micron tolerances. In order to achieve this degree of polishing precision, the effects of the polishing operation on the work surface must be carefully monitored, and the operation of the ring polisher or other surface finishing machine must be adjusted if the profile of the work surface deviates from the desired profile.
The monitoring and adjustment operations are laborious, iterative, and time consuming. First, at least one workpiece (hereinafter a “sample workpiece) must be removed from the ring polisher for testing. Then, in order to obtain an accurate measurement of the steady-state profile of the sample workpiece, the machine operator must wait a period of time after removing the sample workpiece from the ring polisher to allow the workpiece to reach dimensional equilibrium. This equilibration period depends on the material and dimensions of the element but may typically vary from 5 minutes to an hour or more. This delay results from the fact that a sample workpiece is initially distorted with respect to its isothermal shape when it is removed from the ring polisher because of 1) thermal expansion from heat generated by friction during the polishing process and 2) thermal distortion due to liquid evaporation from the sample workpiece. Then, if the workpiece's work surface is more concave or convex than desired (hence indicating that the lap is more convex or concave than desired), the operator must adjust the radial position of the conditioning tool relative to the lap by an amount estimated to compensate for this defect, place the sample workpiece back into its workstation, and wait for the workpiece to be polished for a period of time which is sufficient to permit the effects of the finishing process adjustment to be reflected on the work surface of the workpiece. This period again varies, but may typically be as long as two hours. The sample workpiece then must be removed from the ring polisher once again, allowed to reach dimensional equilibrium, and measured to ascertain the effects of the previous adjustment on the finishing process. The period between the time that a workpiece is initially removed from the ring polisher for topographical measurement to the time that an operator is apprised of the effects of adjustments resulting from that measurement on the polishing operation may be three hours or more.
Even the most skilled operator cannot predict with certainty the effects of many typical adjustments to finishing machine operation. Adjustments typically undershoot or overshoot the desired effect so that, for instance, a workpiece that was initially too convex upon initial testing may be too concave after the adjustment and subsequent testing. As a result, measurement and adjustment must be performed iteratively. Ten to twenty iterations may typically be required for the polishing of a precision polished glass workpiece designed for use as an optical quality element. Since each iteration may take three hours or more, the aggregate period for the polishing process may be thirty to sixty hours or even longer. As most pre-finishing operations such as grinding are capable of producing surface flatness errors on the order of a few microns, and finishing removal rates are on the order of 1 micron per hour, only a small percentage of the finishing period—typically about two to three hours—is required to obtain the desired smoothness. Hence, a polishing process could take as little as two to three hours if it never required adjustment or if adjustment feedback could be obtained instantaneously instead of taking ten to twenty multiples of that time or more.
The above-described iterative polishing process may in some instances be accelerated by using a so-called “monitor plug” or “witness sample” for testing purposes. A monitor plug typically is a workpiece that is very stable and that is not as sensitive to temperature changes as the other, “true” workpieces being simultaneously polished in other workstations. If the monitor plug is held to the desired profile with some specifications, the operator can obtain some assurance that the “true” workpieces in the other workstations have essentially the same profile by periodically measuring the profile of the monitor plug. As a result, there is no need to measure every workpiece every time. However, an iterative measurement and adjustment process still is required, and even the monitor plug must be allowed to reach dimensional equilibrium during each measurement cycle before its profile can be measured.
The above-described iterative polishing process could be accelerated dramatically if profile measurements could be obtained in-situ (i.e., as a workpiece is being worked on a ring polisher or other surface finishing machine) because there would be no need to remove the workpiece from the surface finishing machine and allow it to reach dimensional equilibrium before measurement.
One technique for in-situ metrology is disclosed in “Extending the Accuracy and Precision of In-Situ Ultrasonic Thickness Measurements”, Dunn and Cerino, American Society for Precision Engineering, 1995 Proceedings, Volume 12 (the Dunn and Cerino paper). The Dunn and Cerino paper proposes ultrasonic thickness gauging to achieve precision management of material removal. A plurality of ultrasonic probes are mounted
Caufield Francis J.
Font Frank G.
Lee Andrew H.
Zygo Corporation
LandOfFree
In-situ metrology system and method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with In-situ metrology system and method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and In-situ metrology system and method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2579923