The Advanced X-ray Astrophysics Facility (AXAF) consists of a nested set of six Wolter Type 1 x-ray telescopes. Each telescope consists of two mirrors (a parabola and a hyperbola). The high resolution optical performance required by AXAF and the size of the mirrors necessitates enormous quantities of data to characterize the optics. We will describe an end-to-end data system to be used for the metrology and fabrication of these 12 mirrors. The data system must have the capability of collecting optic metrology data from several instruments, processing and analyzing data, and generating machine instructions for the next grinding or polishing cycle. This system consists of personal computers interfaced to metrology instruments for the automatic collection of data, personal computers that control grinder/polishers, a mainframe computer for storing and managing data, and workstations for data processing and analysis. All of these computers are networked together to facilitate data transfer between computers. The system also includes an extensive library of software whose functions include processing mechanical and interferometric measurements, fitting polynomials to the data, performing frequency analysis of the data, and doing performance predictions. This data system has been used in the fabrication of the first two AXAF mirrors, produced by Hughes Danbury under contract to TRW. These mirrors are the first in a telescope that will be well beyond the performance of any existing x-ray telescopes.

An optical design study for a next generation infrared space telescope has been performed. The concept is that of a passively cooled telescope of minimum aperture 2.5 meter with an F/1.2 primary and wavelength coverage from (lambda) equals 2 to at least 40 micrometers , and possibly to 100 micrometers . Compactness, low thermal emission from the optics and structure, diffraction limited imaging at (lambda) equals 2 micrometers , and sensitivity to misalignment aberrations and manufacturing errors were the main considerations for this study. Ray tracing results are presented showing the characteristics of the various designs considered. A preliminary investigation of stray light properties is also given. Special emphasis has been placed on the testing of such a fast primary, and optical systems using a lateral shearing interferometer are described for testing both the primary and the primary/secondary combination.

New methods have been developed to test telescope optics either at the fabrication stage or while in operation. Results recently obtained at the University of Hawaii are presented.

The use of the Hartmann interferometric optical testing method is described. The method is applied for laboratory testing during mirror manufacturing. Possibilities and advantages of the method in testing mirrors of the 8 m class are discussed. A telescope wavefront sensor based on the method is described and test results of the 2.5 m Nordic Optical Telescope are presented.

The local radius of a reflecting surface can be measured by imaging a grating in a F-Theta- Setup. The radius is determined by measuring the grating period. That can be done in a moire technique where the reference grating is a moire itself. The reticule can be scaled for direct reading of the radius by using the properties of the secondary moire.

We have previously described a new noncontact profiling technique that involves measuring the test surface curvature on a point-by-point basis. Curvature is measured by simultaneously measuring the test surface slope at two slightly displaced locations. As the pair of sensing beams is scanned along the test piece, a profile of curvature is built, from which the height profile is deduced. The sensing of curvature eliminates the need for a reference surface, and makes the approach insensitive to all types of vibration and drift, both in surface height and in surface slope. In this paper, we discuss some of our more recent calibration and measurement efforts in testing steep optics. We also discuss test piece alignment. This aspect of metrology is always of concern, especially in the case of steep optics in general, and steep aspheres in particular. We show that the curvature profiling technique is inherently much less sensitive to unknown misalignments and variations in scanning geometry than a height profiling or slope profiling technique.

Metrology has historically been one of the most formidable hurdles in fabricating large, generalized aspheres. The heart of the problem is that exotic aspheres preclude the classical technique of interferometrically testing from the center of curvature with a spherical beam, driving the need for null correctors. Although the null corrector approach is feasible, the difficulty of certification and the fact that each corrector allows the testing of only a single aspheric form make it expensive and time consuming. The problems are of course much worse for large convex optics, where the center of curvature is not available. Finally, these problems only accentuate the problem of measuring figure during the transition from grinding to polishing. In this paper we discuss a new, non-interferometric instrument currently under development. Non-interferometric techniques are usually not chosen for aspheric metrology, either because they are too inaccurate, involve mechanical contact, or cannot accommodate a large test piece. Our approach is noncontacting and self-referencing, and can be easily expanded from its broadband capacity of 0.5 meter. The instrument consists of a three- dimensional coordinate reference system and a measurement head whose location is tracked within the reference system. The reference system involves no physical reference surfaces, but instead uses an arrangement of laser beams whose positions are monitored with position sensitive detectors. The measurement head uses a new autofocus technique that allows accurate testing even before the surface is polished. Using this head, there is no loss of knowledge during the tricky transition from grinding to polishing. We present here the overall instrument concept, as well as current status and expected performance levels.

We have developed a toroidal surface measuring interferometer system. One of the key features of our interferometer system is its ability to measure a toroidal surface without a toroidal wave front and a toroidal reference surface. Our optical system consists of a Fizeau interferometer using a spherical wave front. A cross section figure of the test surface is obtained to analyze interference fringes like a striped bar by using the Fourier transform method. We can then connect each cross section over the entire surface figure by rotation of a rotary table. Results of examination show that the accuracy of our measuring system is better than (lambda) /20 P-V.

In the field of photonics, there is an increasing need to couple arrays of optical fibers with arrays of receivers or transmitters. One method of doing this is by using micro-lens arrays on the same pitch as the active devices. This paper addresses the important concerns of focal point center-to-center accuracy and the coupling efficiency of these micro-lenses using computer image analysis. Although the image processing techniques used are conventional, their application to micro-lens testing presents a novel approach to determining manufacturing accuracy and lens coupling efficiency. Data on the behavior of the micro-lens can be gathered using the micro-lens array to image a laser beam through a relay lens and onto a CCD camera detector. This paper presents some preliminary experimental results and describes the algorithms that perform the following functions: (1) Using standard Ronchi rulings, calculate image scale and determine if geometric distortion caused by the relay lens is present in image data. (2) Find focal spots of multiple lenses in an image using region aggregation via recursive connectivity evaluation providing sub-pixel moment calculation. (3) Calculate lens coupling efficiency by summing the intensities of all pixels composing the focus spots and comparing them with similar data representing the light incident on the first surface of the lens.

A compact, self-contained production instrument designed to permit the rapid and precise performance characterization of a wide variety of lenses and optical systems has been developed by Eidolon Corporation. The Eidolon Production Nodal Slide/MTF Measurement System can be used to measure effective focal length (EFL), distortion, field curvature, chromatic aberration, spot size, and modulation transfer function (MTF).

Interferometric testing of large optics over long path lengths has been hampered by vibration in the test set-up. The precision of phase measuring interferometry has not been able to provide measurements in vibration environments due to the time required to perform the required phase shift between multiple images of the interferogram. The simultaneous phase shift interferometer (SPSI) has eliminated effects of vibration from phase measurements by creating four separate phase shifted interferograms simultaneously, viewed with four CCD cameras. The CCD cameras provide electronic shutter exposure control which effectively 'freezes' the interference patterns producing high contrast interferograms even with severe vibration. Polarization optics are used to maintain the appropriate phase relationships between the four interferograms. Four separate synchronized video digitizers are used to digitize the interferograms to a maximum resolution of 380 by 240 pixels by 8 bits per pixel. The phase at each pixel in the interferogram is calculated by a PC/486 based microcomputer which also provides complete analysis and graphics of the measurement. Averaging of multiple measurements to reduce the effects of air turbulence is done automatically.

This paper presents the technique of an on-machine precision profile measurement, through a kind of common-path interferometer using a zone-plate. We could analyze profile errors of a spherical mirror on the machine turning at a speed of 900 rpm. The results of experiments are evaluated.

This paper describes the optical design, fabrication and use of single element null lenses for the interferometric testing of generalized aspheric surfaces. The aspheric singlet is designed to work in collimated light, without a field lens, and allows for measurement of surface figure error over the entire clear aperture. The null lens is a single element of multispectral zinc sulfide (ZnS). This material is durable, easily diamond machined and provides good transmission at the test wavelength of 0.6328 (mu) . All reference surfaces needed for alignment are diamond machined into simple collars that hold the null lens and surface under test. This allows easy and accurate optical alignment and minimizes setup time. The procedures for optical layout and alignment will be discussed as well as an error analysis including fabrication and alignment sensitivities. Several examples of the use of this element for testing aspheric surfaces will illustrate the technique. Finally, an extension of the technique to diffractive null optics will be presented.

A technique is described for aligning an autocollimating test of an off-axis paraboloidal segment. A corner reflector (corner-cube retroreflector) is incorporated as an alignment aid to interferometrically align a paraboloidal segment to a spherical wavefront. This alignment task is typically quite challenging and time consuming because the interferograms obtained from a partially aligned autocollimating test cannot be easily interpreted for alignment correction. The normal alignment process requires iterative cycles of adjustment, realignment, and evaluation to achieve results. The use of a corner reflector significantly reduces the time required for the alignment of a n autocollimating test because it provides images and interferograms that can easily be interpreted for alignment correction. The wide dynamic range of this alignment technique makes it applicable to a variety of alignment tasks involving paraboloidal mirrors.

The proper centering of the cutting tool's motion with respect to the spindle axis is essential for the generation of programmed surface figures on turned optical components. The tool centering error, or decentration error, is decomposed into horizontal and vertical displacements. Horizontal decentration imparts a conical figure error to programmed spherical surfaces. Visual interferometric analysis of this figure has been used to estimate the magnitude and sign of this offset. Vertical decentration cannot be estimated by fringe pattern analysis since it does not generate a significant figure error. Here the application of phase measuring interferometry is suggested to obtain automatic estimates of these errors. The horizontal error is obtained by fitting a simple radial polynomial to the phase data. The vertical offset is derived from the extent of a poorly modulated central zone.

This paper describes a method to measure the birefringence of an optical window. The transmitting wavefront includes the contributions from the two surfaces, the material inhomogeneity, and the birefringence. Because of the birefringence, the transmitting wavefront has different profiles for different orientations of polarization of linearly polarized beams. From this difference, the amount of phase difference for the fast and slow axes is obtained. Thus, the birefringence is calculated. With this method, the contributions from the two surfaces and the material inhomogeneity are removed. A laser rod was measured with different methods. The theoretical derivation, comparison of different methods, and experimental results are presented.

Center for Optics Manufacturing (COM) programs have matured beyond the framework of organizational activities. This paper describes COM's current activities and the industry, academic and government support that has made the program successful. The Center, established in 1989 at the University of Rochester's Institute of Optics, is a collaborative effort supported by American Precision Optics Manufacturers Association (APOMA), several academic institutions (Arizona, Central Florida, and Rochester) and the U.S. Army Material Command. The objective is to introduce innovative techniques and advance processing technology that will increase the competitiveness of the optics manufacturing base. Industry, academic, and government cooperation and support has been exceptional.

The Army Manufacturing Technology (MANTECH) program has exerted a strong and lasting influence on the U.S. defense industrial base for more than 25 years. The defense industrial base can be described as the aggregate ability to provide the manufacturing technology, research, development, and resources necessary to produce defense material. The Army's MANTECH program is designed to develop manufacturing thrusts (such as optics) that will result in the development of new and improved manufacturing processes. The Joint Logistics Commanders' assessment of the precision optics industrial base emphasized the need to modernize optics fabrication techniques and equipment. As a result of the study, the American Precision Optics Manufacturers Association (APOMA), the Department of the Army, and several academic institutions have collaborated to establish and support the Center for Optics Manufacturing (COM) at the University of Rochester. Through this successful collaboration of knowledge and resources, participants are able to leverage and reduce the cost, risk, and difficulty of developing new automated optics manufacturing systems and supporting technologies. The optics manufacturing thrust is now in its second year and support from industrial, academic, and government participants continues to increase.

The Center for Optics Manufacturing (COM) has organized a volunteer Process Science Committee that will cooperate in advancing the optical manufacturing sciences. The objective is to develop technical information and processes that improve manufacturing capability, especially in grinding and polishing technology. Chaired by Donald Golini of Litton Itek Optical Systems, the committee members are volunteers from several American Precision Optics Manufacturers Association (APOMA) companies and institutions. Many of the companies are also funding project elements. The committee will accelerate industry progress by integrating the research and development activities of cooperating APOMA companies and institutions involved in both COM and independent programs. In the short term, the effort concentrates on grinding and polishing process innovation. In later phases, the effort will aid in the design future generations of machines and processes. While the developments are directly adaptable to COM's OPTICAM program, the results will influence a wide range of innovation and application in all methods of optical fabrication. Several leaders in the field are participating in the research and development effort--Boston University, Eastman Kodak Company, Hughes Leitz Optical Technologies, Lawrence Livermore National Laboratory, Litton Itek Optical Systems, Melles Griot, Optical Components Inc., Precision Optical, Rank Pneumo, Schott Glass Technologies, Solution Technology, Texas Instruments, Tropel, and the universities of Arizona and Rochester. Other APOMA member companies will participate as resource needs grow. The collaboration is unique in the industry's history.

Optical design, engineering, and manufacturing operate as independent entities. Outmoded specifications for material, geometry, tolerances, and mounting add to cost, lead time, and manufacturing complexity of both military and commercial optics. The optics industry maintains outdated stand-alone design, engineering, and manufacturing systems that do not support integration or communications. This single island of technology adds greatly to the final cost of optical systems.

The ion figuring process has been successfully used to correct the residual surface figure error on a 1.8 m ZerodurTM off-axis segment of the new Keck telescope primary mirror. The segment is one of 36 hexagonal mirror segments composing the full aperture of the 10 m primary mirror. Ion figuring is an optical fabrication method that provides highly deterministic error correction of previously polished optical surfaces using a directed, inert, and neutralized Argon ion beam to physically sputter material from the surface. Figure error correction is accomplished by varying the velocity of the constant-output ion source as it scans across the optic surface. The surface figure error was reduced from 0.726 micrometers rms to 0.090 micrometers rms in two test-figure iterations. The demonstration provided information and requirements for future processing of ZerodurTM and other glass-ceramic materials, and clearly showed the applicability of ion beam figuring to the final correction of large, complex optics.

In this paper we consider an approximation theoretic approach to the computer numerically controlled manufacturing of optical studies. Given the kinematic parameters defining a grinding/polishing tool, we develop an expression for the associated material removal rate. Knowing this, and assuming descriptions of the tool center path and speed, we can then derive a general formula for the amount of material removed at a point on the workpiece by the machine. The final phase of the analysis centers on the determination of strategies for tool center movement so as to achieve a desired pattern of material removal. We discuss two means of formulating such a strategy. The first involves the use of constrained best discrete Lp approximation problems, which for p equals 1,2 can be solved by linear and quadratic programming methods. The second employs the notion of a mollifier or smoothing function and avoids the need for discretization. The results of some computational experiments based on the above methods are included.

Rank Pneumo has worked with the Center of Optics Manufacturing to design a multiple-axis flexible machining center for spherical lens fabrication. The OPTICAM/SM prototype machine has been developed in cooperation with the Center's Manufacturing Advisory Board. The SM will generate, fine grind, pre-polish, and center a spherical lens surface in one setup sequence. Unique features of the design incorporate machine resident metrology to provide RQM (Real-time Quality Management) and closed-loop feedback control that corrects for lens thickness, diameter, and centering error. SPC (Statistical Process Control) software can compensate for process drift and QA data collection is provided without additional labor.

The topography of a smooth machined silicon flat has been measured with a phase-shifting Linnik microscope and a Talystep mechanical profilometer and the results compared in the frequency domain. Excellent agreement is found for the strong low-frequency components that determine the gross properties of the surface. Differences observed for some small high- frequency components may have implications for the understanding of microscope imaging properties.

This paper presents a general method for measuring the frequency response of a surface profiling instrument. One important consequence of this work is the discussion of the importance of the modification of surface spatial frequencies by the measuring instrument. The method uses a step height sample to characterize the impulse instrument, and power spectrum response. Experimental measurements are shown for a variety of focusing objectives. The results show that the half power bandwidth provides a good characterization of the high spatial frequency limit of a surface profiler.

The stressed-lap polishing technique has been developed to meet the challenge of polishing 8- m-class mirrors with highly aspheric figures to an accuracy consistent with the best ground- based telescope sites. The method is currently being demonstrated in the polishing of two primary mirrors, a 1.8-m f/1.0 ellipsoid and a 3.5-m f/1.5 paraboloid. The figure accuracies achieved at the time of writing are 43 nm rms surface error for the 1.8-m mirror, and 190 nm rms surface error for the 3.5-m mirror. Polishing is proceedings on both mirrors. In this paper we describe the process used for the 3.5-m mirror and the progress through the early stages of fabrication. We also summarize progress on the 1.8-m mirror.

A quick and inexpensive method for characterizing spot tools in an optical shop environment is presented. The interpretation of results is explained. Corrective action is suggested.

In this paper, we design a new double-sided lapping method that can accurately and efficiently lap extremely hard sapphire plates. We use a chemo-mechanical polishing method for the final polishing to get a scratch-free surface.

We have developed a new birefringent common-path interferometric method incorporating an automatic data processing system for testing large convex spherical surfaces. This interferometer can eliminate system errors and achieve accurate results of measurement. Precision is about (lambda) /50 and accuracy is about (lambda) /10.

This paper describes a method for measuring MTF of microscope objective, especially high- power oil-immersion microscope objective. The spatial frequency that can be measured is up to 4000 c/mm. The measurement principle, apparatus construction, and measurement results are presented. The results show that the apparatus can be applied to the image evaluation and quality test of microscope objective.

In this paper, we describe an optical method of measuring the reflectance and absorptance of metals by means of a self-made apparatus.

The aspheric surfaces can be tested by using interference type computer-generated holograms (CGH). However, the measurement results are affected by the system errors that exist in optical arrangement. In order to improve the test accuracy, a quantitative method for processing the system errors has been investigated. Experimental results show this method is feasible and effective.

Methods of shaping aspherical surface by grinding and polishing the workpieces are of the greatest interest in manufacturing optical components. One of the most conveient method is to use traditional loose abrasive grinding machine with local figuring pitch. Real time simulation of the aspheric loose abrasive grinding process base on a modified Preston equation and correlated with the shape of local figuring pitch is built up. The simulated target is basically on two spindle grinder'to generate axis symmetric aspherical. The cornputer simulation parameters include the frequencies of the tool and workpiece, the length of oscillating arm, the frequency of eccentric osillation, and historical effect ets. The surface profile related to the grinding factor vs. Workpiece diameter are drawn on the displayin realtime. The sucessive changes of surface profile is shown on display, the shape of analysis of the local figuring pitch are also dicussed in this paper. The results are close to our simulation result.

The Advanced X-ray Astrophysical Facility (AXAF) is the third components of NASA's Great Observatory Program, following the Hubble Space Telescope and the Gamma Ray Observatory. It will be used to study high energy astronomical phenomena in the x-ray spectrum from 0.1 - 10 KeV (124 - 1.24 angstroms). The optical assembly is made up of a nested set of Wolter Type I x-ray telescopes, each consisting of a confocal parabola/hyperbola pair. The 12 near cylindrical, grazing incidence elements are 1 meter in length and range from approximately 0.5 - 1.2 meter in diameter. The instrumentation for circumferential metrology of the AXAF optics uses a standard laser displacement measuring system in order to make precision measurements over a large dynamic range. A unique requirement of this laser system is that it must rotate on a precision bearing during data acquisition. A major error source minimized during the design of the instrument is the thermal load introduced by the laser head. In the final design, a series of quarter wave plates is used to decouple the laser from the rotating optical system while at the same time maintaining the desired polarization states of the incident laser beams. These polarization optics also introduce an angularly dependent path length error into the measurement system that, uncorrected, exceeds the requirements for the metrology system as a whole. This paper describes the details of the system, as well as the calibration method used to eliminate this error source.

The polishing removal rates of fused silica under 20 test conditions have been characterized by the resulting Preston coefficients and an inspection of surface quality using a Nomarski microscope and an optical heterodyne profilometer. Preston coefficient values of > 16 X 10-14 cm3/dyne-cm have been established, approximately a factor of 10 higher than cited previously. However, operational variables can reduce removal rates significantly. Process parameters are shown to have an effect on surface quality that has been related to surface deposits. The effects of chemical additives on removal rates were of special interest. Of importance here is not necessarily the introduction of such additives in practice. The results may improve our understanding of the chemistry of polishing.