This PDF file contains the front matter associated with SPIE Proceedings Volume 12222, including the Title Page, Copyright information, Table of Contents, and Conference Committee listings.

Touch-probing with a Coordinate Measuring Machine (CMM) is not new but contact-measuring a sensitive optic for use in space flight or other vacuum applications is usually considered high risk and avoided at all costs due to specialty substrate materials, optical thin film coatings, and tight surface error tolerances needed for high performance systems operating at challenging wavelengths. In an environmentally controlled cleanroom with a CMM, we inspect the surface damage from touch-probing a variety of optics for use in space flight missions. Motivation comes from the requirement to both characterize an optic and its coordinate system for use in complex, opto-mechanical alignments with single-digit micron accuracies. Currently, a multi-step/instrument process is performed to prevent surface damage, relate the optic’s reference frame to metrology targets on a mount or other associated hardware, and then confidently track the optic’s orientation throughout integration and test. Disadvantages of this measurement combination include error stack-ups, hardware-handling safety, increased exposure to contamination, multiple instrument availability, personnel logistics, and extended schedules. We report on experiments with techniques to mitigate these risks, to create a catalog capturing the measurement parameters used on each space-qualified substrate and coating, and to show surface damage results on the order of single-digit nanometers after touch-probing. Until non-contact, continuous-measurement, multi-axis probes with high accuracy exist, this touch-probing technique shows promise for absolute metrology on sensitive, space flight optics by reducing the risks of conventional multi-step/instrument processes.

The Ocean Color Instrument (OCI), which will be integrated with the Plankton, Aerosol, Cloud, ocean Ecosystem (PACE) satellite, will collect science data that will be used to monitor the health of Earth’s oceans and atmosphere. The Short-Wave Infrared (SWIR) Detection Assembly (SDA), built and characterized by Utah State University Space Dynamics Laboratory (SDL), is a subsystem of OCI consisting of 32 channels covering seven discrete optical bands of interest. A total of 16 SWIR Detection Subassemblies (SDSs) compose the SDA and house the cold optical system. The science data optical input for each SDS is supplied by a 0.22 NA multimode fiber interfacing with a fiber adapter. The diverging light from the fiber is collimated, split by a dichroic beamsplitter to two separate channels, filtered by the science filter, and then reimaged onto the single-element detectors with a final 0.76 NA. Aspheric, diamond-turned powered elements are used throughout the optical design. Fabrication and alignment tolerance analysis/budgets are balanced to ensure the optical system meets throughput requirements. All systems are aligned at ambient temperature using an InSb camera and an in-line illumination microscope system to directly image the active detector area through the science filters. Compensators used during alignment are detector focus and decenter, which are adjusted via photoetched shims in increments of 25 μm. Average focus and centering errors were less than 8 μm among all 32 flight and 10 flight spare detectors. Each SDS spectral response and conversion gain was verified at operational temperature of -65°C in vacuum.

The Point Source Microscope (PSM) is used to find five aberrations related to the symmetries of the autostigmatic image viewed when aligning aspheric mirrors to a point along an axis. These five aberrations exactly match in number the five degrees of mechanical freedom required to align the mirror to an axis and thus provide an exact solution to a unique focus and alignment to an axis. We show how the PSM is used to capture and analyze a set of images as the PSM is moved through focus using the symmetry properties of the image.

We have proposed a new alignment method which is the combination of deflectometry and the sine condition test. One of the great advantages of the new approach is that we need a camera and an LCD monitor larger than the clear aperture of the telescope instead of an interferometer and a return flat. To determine the state of the alignment, we have to place the monitor at two different locations: ideally at the rear principal plane of the telescope and a few meters displaced from the rear principal plane. However, for practical reasons, we may have to place the monitor closer to the telescope. We have simulated how changing the monitor location impacts the alignment, and we show the consequences of variations in the LCD locations on the alignment of a telescope using the new method.

Quality of focus is a critical requirement for structured light illumination (SLI) systems. We present further development of the quad target method, an alignment technique to linearize defocus for SLI systems, by substituting a flat target continuously toggling between two positions. The flat target method enables granular near-real time projector focus assessment with a more universal target and mitigated errors compared to the quad target method.

The 5D-alignment of multi-element assemblies within a single barrel by adhesive bonding represents a powerful alternative to the sub-cell mounting concept. A bifocal autocollimation head offers efficient and simultaneous monitoring of the single element’s tilt and shift within the cell and minimizes the impact of error propagation. The described process allows for production of mounted lenses with an outstanding alignment precision below 1 μm in all dimensions in less than 3 minutes per lens element. The automated process greatly simplifies the transfer of existing, conventional optomechanical designs to higher performance independent of the number of lens elements in the assembly.

A new system has been developed to improve the accuracy of aligning a laser beam towards a specific location on a grid board. The system consists of automatic laser spot detection and spatial calibration, both in real time. Laser detection uses OPEN-CV libraries in Python to get precise coordinates of laser spot contour center. Comparison with coordinates from point to aim, gives real time deviation and allows the operator the adjust laser direction to meet the quality criterion. This system improves overall alignment uncertainties by a factor of 3 compared to the classical alignment procedure thanks to precise laser spot location and helpful assistance in in setting the laser direction.

Passively athermalized optical systems produce high quality images over a large thermal range without actively adjusting focus. This athermalization is achieved through careful selection of the glass for each lens and metal for each mount. For drop-in systems, the material combination for best optical performance often leads to a lens stack with an overall coefficient of thermal expansion (CTE) that is different from the CTE of the barrel that holds it together. Since bulk glass and metal are relatively stiff, this CTE mismatch results in large variations of the preload force retaining the lens stack in compression over the optical system’s survival thermal range. For this reason, compliant spacers are commonly added to the lens stack in an effort to attenuate these preload force variations. However, the effect of these compliant spacers on the athermalization of optical systems is seldom analyzed. We perform a first-order calculation of the effective CTE of compliant spacers to assess their impact on optical performance and introduce an optomechanical design approach to reduce the amount of compliance needed by matching the overall CTE of the lens stack to the CTE of the barrel.

MTF tests, one of the most important optical metrology tasks for AR/MR glasses, analyze the DUT’s (Device Under Test) optical resolution to provide quantitative feedback for design verification and manufacturing process control. Due to the immaturity of the whole design/fabrication technical chain, current diffractive AR glasses show strong angular resolution non-uniformity across the FOV, and the measurement’s angular accuracy and consistency significantly impact the test repeatability and reproducibility. This paper presents a novel optical calibration apparatus to enable absolute angular alignment/verification, which can be implemented with a small volume and easily fit into the metrology equipment.

This paper first reviews the practical optical calibration constrains of the optical metrology equipment Periscope, which measures AR/VR glass’s binocular disparity by providing the numerical measurement of non-parallelism between two eyes’ optical axes. A self-adaptive calibration methodology with close loop feedback to track on the calibration tool’s accuracy is proposed to precisely calibrate the parallelism of the periscope’s two optical sensing channels and efficiently verify this parameter periodically with consistency over time or instruments. A few implementation schemes, including a passive target, an active selfreferenced target, and two different sensitivity enhanced targets are discussed in depth to compare the performance contributors: accuracy, repeatability, and system complexity, which leads to the recommendations for different application scenarios. Beyond the AR/VR disparity measurement a potential application based on the same methodology is introduced to evaluate the precision motion system’s accuracy and tolerance.

External Fabry-Perot resonators are widely used in the field of optics and are well established in areas such as frequency selection and spectroscopy. However, fine tuning and thus most efficient coupling of these resonators into the optical path is a time-consuming task, which is usually performed manually by trained personnel. The state of the art includes many different approaches for automatic alignment, which, however, are designed for special optical configurations and cannot be generalized. However, these approaches are only valid for individually designed optical systems and are not universally applicable. Moreover, none of these approaches address the identification of the spatial degrees of freedom of the resonator. Knowledge of this exact pose information can generally be integrated into the alignment process and has great potential for automation. In this work, convolutional neural networks (CNNs) are applied to identify the sensitive spatial degrees of freedom of a FabryPerot resonator in a simulation environment. For this purpose, well established CNN architectures, which are typically used for feature extraction, are adapted to this regression problem. The input of the CNNs was chosen to be the intensity profiles of the transversal modes, which can be obtained from the transmitted power behind the resonator. These modes are known to be highly correlated with the coupling quality and thus with the spatial location of resonators. To achieve an exact pose estimation, the CNN input consists of several images of mode profiles, which are propagated through an encoder structure followed by fully-connected layers providing the four spatial parameters as the network output. For training and evaluation, intensity images as well as resonator poses are obtained from a simulation of a free spectral range of a resonator. Finally, different encoder structures including a memory efficient, small self-developed network architecture are evaluated.

In this paper, we describe a successful method to power-on VCSELs while performing active dry-alignment with optical coupling structures on PICs. Dry-alignment is here intended as an optical coupling approach with no temporary or permanent bonding of coupled parts together. Emphasis has been done on achieved results showing low insertion losses (typ. <3dB) obtained when coupling single-mode 1.55um InP VCSEL light beams on top of up-reflecting mirrors (URMs) realized with VTT 3um SOI Silicon Photonic (SiPh) platform. The described method demonstrates a dramatic coupling efficiency increment, overwhelming de-facto most of the drawbacks of the current industry-standard approach when using passive alignments. VCSELs can be tested and replaced several times without any visible damage. A limiting factor such as off-axis errors in VCSELs when coupled with URMs, will be greatly compensated with active dry-alignment. VCSELs´ assembly time on PIC is expected to be 3-5 times longer than with the passive alignment.

Certain companies particularly those with strong design and optical manufacturing units keep strict but private statistical records regarding optical manufacturing and also use the data for design purposes. Design houses without manufacturing sectors are at a disadvantage. However, there is a small but growing public body of knowledge regarding these statistics. In this work, we develop a process to go from gathering raw manufacturing data to using the data for lens system tolerancing. We will describe a tolerancing practice using CODE V and our existing data with the goal of improving our ability to predict manufacturing outcomes. We present reasonable parameter values for the truncated normal and other distributions for variables such as wedge. In some cases, we will link the parameter values to standard tolerance categories that many manufacturers give: commercial, precision, and high precision. The tolerancing method will use established CODE V practice as well as macros with parameter values derived from data as one of the inputs. An example of how to use these techniques on a lens design will be given. We also provide an appendix that classifies glass type by manufacturability. The data provided in the appendix can be used in the tolerancing process.

Optical tolerance analysis is a very important step in optical systems development. It ensures that appropriate optical performances will be achieved considering all the manufacturing errors involved in the assembly. To perform an accurate tolerance analysis, a realistic optomechanical tolerance model and appropriate perturbations simulation are required in the optical design code. Most of the time, optomechanical lens mounting is not taken into account accurately in classical optical tolerancing method. To improve optical system tolerancing process, an integrated opto-mechanical tolerance analysis is proposed. This paper first describes typical tolerancing process and iteration performed between optical designers and optomechanical engineers in the development of optical systems. Then, the optomechanical tolerance analysis that involves interactions between lenses and mounts, as well as manufacturing errors is presented. Simulation methods to consolidate optical and optomechanical tolerance analysis are discussed. Finally, an integrated optomechanical tolerance analysis is described, and a new optomechanical tolerancing software is introduced. The intent of this new modeling method is to perform accurate optical simulations that are representative of the optomechanical mounting and centering methods. This result in a more efficient allocation of the tolerances and a more accurate prediction of the optical system performances.

An innovative software application for a more realistic tolerance analysis has been developed recently by INO. The application is using optical and mechanical manufacturing databases as well as several equations to translate realistic manufacturing tolerances, optomechanical mounting interfaces, and centering methods into tilts and decenters perturbations, easily transferable to Zemax OpticStudio. The standalone application can be used by the optical designer to quickly verify the feasibility of a mounting and alignment technique according to the specific sensibilities of the current design. The optomechanical engineer can also easily validate or choose a better centering method as well as update the mechanical tolerance parameters. Once the parameter is fixed, the optical designer can export the new parameters into a Zemax OpticStudio file, updating Lens Data Editor and Tolerance Data Editor. Communication between both optical and optomechanical specialists is straightforward with this powerful tool, making the design process easier, quicker, and more accurate. This paper presents how INO is using its standalone application for tolerance analysis to overcome the complex simulation of various centering techniques. Through real examples, it will show how realistic tolerancing simulations impact the choice of appropriate centering method for the lens assembly.

Designing fast and high-resolution 3D measurement systems, which are key enabler for the production of the future, is a challenging task, since optical, electrical and mechatronic components with respective specifications need to be integrated. Particularly the resulting measurement uncertainty is of highest interest but can currently only be roughly determined in advance. This paper proposes an uncertainty framework for an optical scanning 3D triangulation sensor system to calculate the influence of the alignment and component specifications on the achievable system performance. This enables to calculate the required uncertainty of each component for given overall uncertainties in the lateral and axial direction. The default and best achievable specification for the manufacturing tolerances of each component, sensor noise and resolution of the detector, and angular resolution of the fast steering mirror used to manipulate the illumination path, are specified in advance. To keep the overall system cost low, the simulation of the optical path is initially performed with the default specifications. By comparing this simulation result with the ideal case, the overall uncertainties and contribution of each component can be determined. If the calculated uncertainties do not meet the requirements, the specification for the component, which contributes most to the uncertainty, can be gradually improved within the maximum specification. The procedure is repeated until the required levels of uncertainty are obtained or until it cannot be further improved since the maximum specifications have been exceeded. This ensures that only the specifications required to achieve the specified uncertainty are tuned.

We study the propagation of wavefronts refracted through separated doublet lenses (SDL), considering a plane wavefront propagating parallel to the optical axis. We provide formulas for the zero-distance phase front refracted through SDL by using Huygens’s principle. Additionally, we obtain formulae to represent the shape of refracted wavefronts propagated at arbitrary distances along the optical axis, as a function of all parameters involved in the process of refraction. Finally, some examples for commercial SDL showing the evolution of the wavefronts arbitrary distances are presented, assuming different wavelengths for the refractive indices of the lenses, displaying dispersion effects produced through SDL.

Adaptive optics(AO) compensates for aberration between light detector and imaging target, for example, air turbulence, misaligned optics, and eye lens. This technique, initially developed to improve the performance of astronomic imaging, is leading to advances in the ophthalmic imaging field through combination with various optical imaging systems. General hardware-based AO systems need additional optics and beam paths for adjusting the beam size at the deformable mirror(DM) as well as matching the image plane. Previous research has composed the system using spherical mirrors instead of the lens to reduce aberration. Additionally, previous papers have reported off-axis type AO imaging systems for eliminating astigmatism and more compact equipment than on-axis type. This study optimized an off-axis AO-region of the AO-scanning laser ophthalmoscope(SLO) through optical design by Opticstudio® and three-dimensional rendering by SOLIDWORKS®. The rendering process, including virtual mounting, allows checking whether block or pass of the beam by optomechanics and providing relative coordinate of mounted optics. The verification method for checking the alignment of the system is the comparison between simulation and practical beam wavefront at a specific beam path.

In this paper, we suggested an alignment mark measurement system for die-to-wafer bonder. The system consisted of a vision system to measure the positions of alignment marks on die and wafer opposite to each other, and a height measurement system to detect the heights of the die and wafer to the vision system. Besides them, a tilt measurement system was also attached to check the parallelism between the die and wafer for exact measurement and bonding. For high precision measurement, the vision system used a special prism structure which minimized the distance between the die and wafer and measured alignment marks on both sides simultaneously. A focus tunable lens was also applied to control a focus position without changing the height of the system. We have designed and built the optical system in a compact size and presented some preliminary results here.

Synchrotron beamline alignment is often a cumbersome and time-intensive task due to the many degrees of freedom and the high sensitivity to misalignment of each optical element. We develop an online learning model for autonomous optimization of optical parameters using data collected from the Tender Energy X-ray Absorption Spectroscopy (TES) beamline at the National Synchrotron Light Source-II (NSLS-II). We test several optimization methods, and discuss the effectiveness of each approach, as well as their application to different optimization problems and benchmarks for beamline performance. We also discuss the practical concerns of implementing autonomous alignment systems at NSLS-II, and their potential use at other facilities.