This PDF file contains the front matter associated with SPIE Proceedings Volume 8131, including the Title Page, Copyright information, Table of Contents, Introduction, and Conference Committee listing.

Asymmetric imaging errors are frequently the main cause for tight tolerances and high demands on manufacture and assembly of optical systems. In order to simultaneously increase robustness and reduce manufacturing cost, desensitization strategies can be applied. Tolerance effects have been included into the optimization function (merit function) by some lens designers to find insensitive designs ^1-5 and frequently compensators are employed to further improve the performance of assembled lenses. Compensators are limited to a small number of system parameters, but selective assembly of components can extend the number of parameters available for compensation. It can be employed to reduce tolerance effects of disturbed parameters by finding the best matches out of a set of components. The potential of using tolerance desensitization in conjunction with selective assembly to reduce asymmetric errors in imaging optical systems is investigated. A focus is given on strategies to find tolerance insensitive design forms under the presence of selective assembly compensators and the selection of suitable parameters for desensitization and measurement.

Several methods have been demonstrated for desensitization of a lens design to manufacturing errors with the result of increased as-built performance at the expense of a slightly reduced nominal performance. A recent study demonstrated a targeted desensitization method tuned for the most sensitive lens parameters can greatly increase yield for a known set of manufacturing tolerances. The effectiveness of such a targeted desensitization relies on two key pieces of information; lens sensitivities and manufacturing tolerance distributions. Targeted desensitization to known and unknown manufacturing tolerances is examined with an example demonstrating the impact of designing to unknown, bounded manufacturing tolerance distributions.

Compared to glass optics, differences in the materials, configurations and manufacturing processes of molded plastic optics lead to differences in the approach to their design and analysis. This paper discusses the differences in tolerances and in tolerance analysis between molded plastic optic and conventional glass optic systems.

A Multi-Objective approach for lens design optimization was verified. The optimization problem was approached by addressing simultaneously, but separately, image quality and system tolerancing. In contrast to other previous published methods, the error functions were not combined into a single merit function. As a result the method returns a set of nondominated solutions that generates a Pareto Front. Our method resulted in alternate and useful insights about the trade off solutions for a lens design problem. This Multi-objective optimization can conveniently be implemented with evolutionary methods of optimization that have established success in lens design. We provided an example of the insights and usefulness of our approach in the design of a Telephoto lens system using NSGA-II, a popular multiobjective evolutionary optimization algorithm.

A case study is presented to illustrate some of the performance and cost driving tradeoffs involved in the design and tolerancing of optics and optomechanics, where simple lens barrels are concerned. A double Gauss lens was designed, along with two variations of lens barrel designs for mounting the lenses. The two lens barrel designs are compared, and one is selected for further analysis. Tolerance selection guidelines are given and discussed for the optics, as well as the optomechanics. It is shown how the optomechanical tolerances (axial spacing, element decenter, and element tilt) are derived from their primary parameters for the selected lens barrel design. Finally, Monte Carlo analysis is used, along with the provided tolerancing guidelines, to determine reasonable tolerances in order to satisfy the optical performance and yield requirements.

An experimental test for violations of the sine condition is presented. This test is particularly useful for identifying the state of alignment of an imaging system because it provides a direct measurement of the linear astigmatism (astigmatism that varies linearly with field) in a system using only on axis measurements. The concept of the test is explained from the perspective of both geometrical optics, using the sine condition, and wave optics. In addition, the results of an experimental proof of concept are presented. This experiment shows good agreement between the measured and predicted results.

A novel fiber-optic probe measures the velocity distribution of an imploding surface along many lines of sight. Reflected light from each spot on the moving surface is Doppler shifted with a small portion of this light propagating backwards through the launching fiber. The reflected light is mixed with a reference laser in a technique called photon Doppler velocimetry, providing continuous time records. Within the probe, a matrix array of 56 single-mode fibers sends light through an optical relay consisting of three types of lenses. Seven sets of these relay lenses are grouped into a close-packed array allowing the interrogation of seven regions of interest. A six-faceted prism with a hole drilled into its center directs the light beams to the different regions. Several types of relay lens systems have been evaluated, including doublets and molded aspheric singlets. The optical design minimizes beam diameters and also provides excellent imaging capabilities. One of the fiber matrix arrays can be replaced by an imaging coherent bundle. This close-packed array of seven relay systems provides up to 476 beam trajectories. The pyramid prism has its six facets polished at two different angles that will vary the density of surface point coverage. Fibers in the matrix arrays are angle polished at 8°to minimize back reflections. This causes the minimum beam waist to vary along different trajectories. Precision metrology on the direction cosine trajectories is measured to satisfy environmental requirements for vibration and temperature.

Although the PSM is primarily an alignment instrument, it can also be used to determine the conjugates of parabolic and elliptical off-axis mirrors. By positioning the PSM at the sagittal and tangential foci of the mirrors, the conjugate distances of the mirror can be found using a laser range finder, for example. Knowing the sagittal and tangential radii of curvature (Rs and Rt), the vertex radius (Rv) is easily calculated. This information is used to verify that the mirror has been correctly manufactured and to aid in positioning the mirror in an optical system. Examples are shown of these steps.

We characterize the precision of a low uncertainty alignment procedure that uses computer generated holograms as center references to align optics in tilt and centration. This procedure was developed for the alignment of the Wide Field Corrector for the Hobby Eberly Telescope, which uses center references to provide the data for the system alignment. From previous experiments, we determined that using an alignment telescope or similar instrument would not achieve the required alignment uncertainty. We developed a new procedure that utilizes computer generated holograms to create multiple simultaneous images to perform the alignment. The center references are phase etched Fresnel zone plates that act like thin lenses. We use zero order reflections to measure tilt and first order imaging from the zone plates to measure centration. We performed multiple alignments with a prototype system consisting of two center references spaced one meter apart to characterize this method's performance. We scale the uncertainties for the prototype experiment to determine the expected alignment errors in the Wide Field Corrector.

Optical system design is always constrained by achievable fabrication tolerances, and there is a constant balance between design performance and the cost or yield of the fabrication process. However, many of the best designs are achieved by starting from the manufacturing platform, and modifying the basic structure of the system to take maximum advantage of symmetries in the system. We will describe several optical systems whose symmetries have allowed us to bypass some of the more problematic tolerances. The first is a multi-reflection imaging system using concentric aspheric mirrors, diamond turned into a a single optical element, which allowed us to create a 3x magnification Galilean telescope just 1 mm thick, designed to be incorporated into a contact lens as a vision aid. The second system is a multi-scale lens design which explores a different type of symmetry: a bilateral monocentric primary lens, followed by over 200 identical secondary optics, which together form an aggregate 2500 megapixel imager. And the third system is a non-imaging solar concentrator using micro-optic lenslets and micro-reflectors which couple incident sunlight into a slab waveguide, where the problem of aligning the lenslets to the micro-reflectors has been bypassed by using the focal spot from each lenslet to form it's corresponding injection feature.

Previous papers have established the inadvisability of applying tolerances directly to power-series aspheric coefficients. The basic reason is that the individual terms are far from orthogonal. Zernike surfaces and the new Forbes surface types have certain orthogonality properties over the circle described by the "normalization radius." However, at surfaces away from the stop, the optical beam is smaller than the surface, and the polynomials are not orthogonal over the area sampled by the beam. In this paper, we investigate the breakdown of orthogonality as the surface moves away from the aperture stop, and the implications of this to tolerancing.

The process of assigning tolerances to optical designs is intrinsically statistical, regardless of volume. Papers covering the statistics of tolerancing, however, have been infrequent. This paper will discuss the statistical nature of tolerancing including ramifications that all optical designers and engineers should understand.

This paper outlines methods for dimensioning and tolerancing lens seats that mate with spherical lens surfaces. The two types of seats investigated are sharp corner and tangent contact. The goal is to be able to identify which seat dimensions influence lens tilt and displacement and develop a quantifiable way to assign tolerances to those dimensions to meet tilt and displacement requirements. After looking at individual seats, methods are then applied to multiple lenses with examples. All geometric dimensioning and tolerancing is according to ASME Y14.5M - 1994.

The Large Binocular Telescope on Mt Graham in Arizona consists of two 8.4 m telescopes mounted on a common gimbal. Each independent telescope has hexapods controlling the position of individual optical elements. These can be used to drive each telescope to point to a common target (or known offsets to these) as is required for many of the observational modes of the telescope. The hexapods have a limited range of travel, particularly the primary mirror hexapods. This paper discusses the approach that has been taken to achieve optical co-pointing while maintaining the maximum possible range of travel in the hexapods. The approach described here is, starting with collimated but not co-pointed telescopes, to first calculate a coma-free rotation of the optical elements that will equalize the percentage consumption of range on pairs of hexapod elements that affect {X,Y} pointing; i.e. {X, Ry} and {Y, Rx} respectively. On a collimated telescope this results in a state which maximizes the available range of travel of the hexapods for a given set of initial hexapod values. Next a further calculation step is taken which achieves optical co-pointing. This step takes into account what range of travel is available for each hexapod for the given "range-balanced" starting point, then allocates a percentage of the required optical copointing to each telescope so as to maximize the available range of hexapod travel on each side. This technique has been applied successfully to both the prime-focus and "bent-Gregorian" modes of the telescope.

We describe the use of LIDAR, or "laser radar," (LR) as a fast, accurate, and non-contact tool for the measurement of the radius of curvature (RoC) of large mirrors. We report the results of a demonstration of this concept using a commercial laser radar system. We measured the RoC of a 1.4m x 1m spherical mirror with a nominal RoC of 4.6m with a manufacturing tolerance of 4600mm +/- 6mm. The prescription of the mirror is related to its role as ground support equipment used in the test of part of the James Webb Space Telescope (JWST). The RoC of such a large mirror is not easily measured without contacting the surface. From a position near the center of curvature of the mirror, the LIDAR scanned the mirror surface, sampling it with 1 point per 3.5 cm². The measurement consisted of 3983 points and lasted only a few minutes. The laser radar uses the LIDAR signal to provide range, and encoder information from angular azimuth and elevation rotation stages provide the spherical coordinates of a given point. A best-fit to a sphere of the measured points was performed. The resulting RoC was within 20 ppm of the nominal RoC, also showing good agreement with the results of a laser tracker-based, contact metrology. This paper also discusses parameters such as test alignment, scan density, and optical surface type, as well as future possible application for full prescription characterization of aspherical mirrors, including radius, conic, off-axis distance, and aperture.

For many optical systems the properties and alignment of the internal apertures and pupils are not critical or controlled with high precision during optical system design, fabrication or assembly. In wide angle imaging systems, for instance, the entrance pupil position and orientation is typically unconstrained and varies over the system's field of view in order to optimize image quality. Aperture tolerances usually do not receive the same amount of scrutiny as optical surface aberrations or throughput characteristics because performance degradation is typically graceful with misalignment, generally only causing a slight reduction in system sensitivity due to vignetting. But for a large deployable space-based observatory like the James Webb Space Telescope (JWST), we have found that pupil alignment is a key parameter. For in addition to vignetting, JWST pupil errors cause uncertainty in the wavefront sensing process that is used to construct the observatory on-orbit. Furthermore they also open stray light paths that degrade the science return from some of the telescope's instrument channels. In response to these consequences, we have developed several pupil measurement techniques for the cryogenic vacuum test where JWST science instrument pupil alignment is verified. These approaches use pupil alignment references within the JWST science instruments; pupil imaging lenses in three science instrument channels; and unique pupil characterization features in the optical test equipment. This will allow us to verify and crosscheck the lateral pupil alignment of the JWST science instruments to approximately 1-2% of their pupil diameters.

The PICARD satellite is dedicated to the monitoring of solar activity. It carries several imaging and radiometric instruments. One of them, SODISM, is a high-resolution radio-imaging telescope measuring the Sun diameter and total flux in near UV and visible wavelengths. Along with mirrors, SODISM includes highly reflective filters and attenuators, which generate ghost images. These disturb the Sun edge area, the total flux measurement and also the fine aiming channel. This is compounded with tilt tolerances, which shift and modify the ghosts images. Stray light was studied through ASAP simulation, with broad sources and high order splits. Each path was studied separately, checking its effect on instrument performance and the possible effect of tilts. Some design improvements allowed to reduce the most critical paths, while others, although relatively intense, stood clear from the critical areas. However ground tests and flight results show some residual ghosts, which could not be fully suppressed due to mechanical tolerances. They shall be taken into account by image processing.

In order to perform spectrometric measurements, the Near Infrared Spectrometer (NIRSpec) aboard the James Webb Space Telescope (JWST) needs the ability to select various spectral band widths and split these up into its comprised wavelengths. These functions are achieved by the Filter Wheel Assembly (FWA) and the Grating Wheel Assembly (GWA). The filters of the FWA select a different bandwidth of the spectrum each while the gratings on the GWA yield specific diffractive characteristic for spectral segmentation. A high spectral sensitivity as well as the ability to detect the spectra of various objects at the same time result in high requirements regarding the positioning accuracy of the optics of both mechanisms in order to link the detected spectra to the 2-dimensional images of the observed objects. The NIRSpec mechanism including FWA and GWA will operate at temperature levels below 42K which are established during testing inside of a cryostat. However the alignment and testing of these mechanisms requires a lot of thought since there is very limited access to the item under test within such a device. Alignment needs to be preloaded based on simulations and testing is reduced to optical methods and evaluation of electrical signals. This paper describes the methods used for the various alignment steps, the corresponding tests and their precision of measurement as well as the achieved accuracies in the mechanism performance.

In order to meet both optical performance and structural stiffness requirements of the aerospace Cassegrain telescope, the primary mirror shall be mounted with the main plate by iso-static mount. This article describes of the alignment of the aerospace Cassegrain telescope's primary mirror and iso-static mount by using coordinate-measuring machine (CMM), and the design and assembly of mechanical ground support equipment (MGSE). The primary mirror adjusting MGSE consists of three 3-axis linear stages and point contact platforms, which hold the mirror while avoid the rotated movement when adjusting the stage. This MGSE provide the adjustment of tilt and height for the mirror. After the CMM measurement, the coordinates of measured point on the mirror will be analyzed by the software based on least square fitting to find the radius of curvature, conic constant, de-center and tilt, etc. According to these results, the mirror posture will be adjusted to reduce de-center and tilt by the designed MGSE. The tilt in x and y direction are reduced within 0.001 degrees and the distance deviation from the best fitted profile of the mirror to the main plate shall be less than 0.008mm.

In our earlier study[12], we suggested a new alignment algorithm called Multiple Design Configuration Optimization (MDCO hereafter) method combining the merit function regression (MFR) computation with the differential wavefront sampling method (DWS). In this study, we report alignment state estimation performances of the method for three target optical systems (i.e. i) a two-mirror Cassegrain telescope of 58mm in diameter for deep space earth observation, ii) a three-mirror anastigmat of 210mm in aperture for ocean monitoring from the geostationary orbit, and iii) on-axis/off-axis pairs of a extremely large telescope of 27.4m in aperture). First we introduced known amounts of alignment state disturbances to the target optical system elements. Example alignment parameter ranges may include, but not limited to, from 800microns to 10mm in decenter, and from 0.1 to 1.0 degree in tilt. We then ran alignment state estimation simulation using MDCO, MFR and DWS. The simulation results show that MDCO yields much better estimation performance than MFR and DWS over the alignment disturbance level of up to 150 times larger than the required tolerances. In particular, with its simple single field measurement, MDCO exhibits greater practicality and application potentials for shop floor optical testing environment than MFR and DWS.