The detailed displacement data provided by finite element analysis (FEA) tools must be translated into forms acceptable by most optical ray tracing tools (CODE V specifically). A useful medium for transferring FEA data is the Zernike circular polynomials that many optical ray tracing tools will readily accept as input. However, the translation process is nontrivial, and two specific difficulties are explored in this paper. The first issue involves a coordinate space transformation that is required because the optically relevant coordinate system is not the same as the Cartesian coordinate system typically used in the finite element model. Several algorithms are described to perform this transformation and their pros and cons enumerated. Specifically, comparisons are made between sag based and surface normal (wavefront) based coordinate systems, and it is found that by using the sag equation of the original surface, the accuracy of the data translation can be improved. The second issue discussed is the accuracy of the polynomial fitting process. The loss of orthogonality stemming from undersampling, nonuniform mesh density, and annular surfaces are discussed with potential work-arounds.

This paper presents a theoretical (closed-form) solution for the z-axis surface deformations of a linear, homogeneous, unconstrained and isotropic paraboloidal surface subjected to a 3-dimensional linear thermal temperature gradient and soak temperature change. Previously, an equation for the component of the nodal surface displacement in the z direction has been published. Attaching the z-axis component of the nodal surface displacement to the original surface does not accurately describe the final surface. This work extends the previous analysis and presents a polynomial equation for the corrected surface deformation along the z-axis, as well as, the coefficients for the standard Zernike polynomial describing the corrected surface deformation. Also included is a discussion about z-axis temperature gradients across the paraboloidal surface and how to calculate an equivalent soak temperature change.

Brittle materials such as glass do not possess a single characteristic strength. The strength of the material is dependent on the distribution of cracks or surface flaws. These factors, coupled with the inherent brittleness (cause of catastrophic or rapid failure) mean that extremely conservative design approaches are typically used for optical elements made of glass. Determining a design allowable for glass elements is critical for optical systems using relatively brittle glass types or for optical elements subject to relatively high stress levels. Rule-of-thumb tensile design strengths are typically at 1000 - 1500 psi for nominal glass materials. This neglects the specific glass composition, subcritical crack growth, surface area under stress, and nature of the load - static or cyclic. Several methods to characterize the strength of optical glass are discussed to aid engineers in predicting a design strength for a given surface finish, glass type, and environment. These include estimating fracture toughness for a given glass, predicting inert strength using material test data, and lifetime predictions accounting for static fatigue and cyclic loading. Determining a design strength for a spaceborne optical element is discussed.

A sensitivity evaluation of mounting 100mm optics using elastomer or bipod flexures was completed to determine the relative effects of geometry, structure, material, thermal and vibration environment as they relate to optical distortion. Detailed analysis was conducted using various finite element-modeling methods. Parts were built and the results were verified by conducting brassboard tests. What makes this evaluation noteworthy is the two vastly different approaches, and how they both exhibited athermal properties and minimized optical distortion. Materials were carefully selected while the geometry and structure were optimized through analytical iteration. The elastomeric optical mount consists of 12 equally spaced pads of RTV placed around the circumference of the optic. These pads were sized to maximize stiffness and minimize surface deformations. The surrounding material was appropriately selected in order to contribute to an athermal design. The bipod flexure optical mount uses three flexures cut from a single piece of material. Each flexure is a bipod oriented to comply radially with changes in temperature. This design is monolithic and uses conventional epoxy at the optical interface. The result is a very stiff athermal design. This paper covers both opto-mechanical designs, as well as analytical results from computer modeling and brassboard tests.

High performance optical systems can be very labor intensive and costly to fabricate and assemble by traditional methods. With the advent of modern diamond turning methods and other key technologies, more deterministic methods of manufacturing optomechanical assemblies that require extremely tight tolerances have been developed. High numerical aperture lenses that must approach diffraction-limited performance at deep ultraviolet (DUV) wavelengths represent some of the most difficult optical systems to manufacture. Modern diamond turning equipment and advanced assembly techniques have been applied to produce a DUV microlithographic lens assembly that features both precision machined subcells and internal compensators for the correction of residual optical aberrations. The details of this process are described in this paper

Finding the components of an optical system that are most sensitive to misalignment allows a designer to insulate them from outside perturbations or incorporate compensators to account for alignment errors. At Advanced Optical Systems, Inc. (AOS) we used opto-mechanical constraint (OMC) equations to analyze the misalignment sensitivity of an optical correlator system and develop a better design. The OMC equations provide sensitivity coefficients for each element in the design that can be used to determine which components create the greatest image shift and focus errors when not optimally aligned. The OMC analysis model of the optical correlator was verified using a test bench with lenses in adjustable mounts to induce known amounts of misalignment in multiple axes. The experimental data matched the calculated values for each tested lens. The OMC coefficients assisted in identifying (1) lenses that are sensitive to loose manufacturing tolerances, (2) where subsystem designs can be beneficial, and (3) materials that provide optimum thermal performance. We will show results from our latest optical correlator package built using the OMC model analysis, which was critical to making decisions in the opto-mechanical design state of system development. We will also discuss a MATLAB simulation of AOS' optical correlator that incorporates the opto-mechanical constraints into a digital simulation of the correlation image.

We investigate a self-aligning method used to couple a vertical emitting laser (flip-chip) to a planar single-mode waveguide through a 45° mirror based upon solder self-alignment. The alignment tolerances to achieve targeted coupling loss of 3.5dB or better were determined for all axes by modeling the optical behavior of the vertical waveguide/45° mirror interconnect. Simulation models for optical design are carried out using commercial software "FullWAVE" from Rsoft, Inc. The design and optimization of the joint's parameters are performed using a public-domain software "Surface Evolver" for surface energy minimization in conjunction with GE proprietary Six Sigma regression and optimization tools. The parameters that were considered for the models included the misalignment along all axes, solder volume, the height of the joints, the radius of the metallization pads, the initial placement error, and the solder reflow time. It has been assumed that intermetallic growth at the solder/metallization/air triple line does not couple with the self-alignment process in order to simplify this problem. The study shows that, given good control over noise parameters such as vibration and reflow temperature fluctuations, solder self-alignment can be harnessed to achieve the targeted coupled efficiency.

This paper presents the development of the IASI Infrared Spot Scan test equipment, with a focus on the mechanical design. The IASI instrument, developed by Alcatel, is a spaceborne meteorological instrument, for observation of the Earth atmosphere in the infrared wavelength region. An infrared Optical Ground Support Equipment (OGSE), developed by TNO TPD is used to test the focal plane of the IASI instrument. The characterization is done by response measurement of an infrared spot scanning the detector area of IASI. The vacuum part of the OGSE consists of 3 linear stages, an optical table comprising an infrared source, an elliptical mirror, a shutter and a diaphragm. The system is partly cooled. A control system for stages, shutter and thermal control completes the system.

Although the possibility of a 6 degrees of freedom adjustment based on a single body pulled onto on six adjustable supports follows directly from the kinematic theory, such mechanisms are seldom used in actual products. Two major drawbacks for the use of this solution are: 1. Due to the sliding contact between the body and the supports, friction will occur and may inhibit movement. 2. Coupling between the adjusted axes cannot be avoided, this may interfere with the necessity of an adjustment procedure with a limited number of iterations This paper presents a matrix calculation method that offers a prediction whether the body will move as required, depending on the position of the supports and on the magnitude of the friction. This method enables to check the functionality of a design. This method has been used in the design of several adjustment mechanisms consisting of a body pulled onto six supports. The matrix calculation method also allows predicting the movement of the adjusted body due to adjustment of the separate supports. Using this it is relatively simple to simulate the movements that an operator will observe, and in this way check whether an operator is capable to handle the couplings present in the adjustment. Using the simulation the adjustment procedure can be optimized.

Measuring the wind speed from a satellite is not new. However measuring with great precision is by far not trivial. Various methods are available for that. A common method is to use the Doppler effect. A UV-laser on board of the satellite is used to "fire" to the earth atmosphere. Some photons will be reflected back to the satellite. Because of the speed of the particles in the air the photons will experience a small Doppler shift. Wind speeds of 1 m/s are hereby equivalent to a wave length shift of 1 femtometer. The paper presents the patented method of how to measure these small wavelength shifts without running into trouble concerning the mechanical design. It will understood that such instrument will be very sensitive to thermal variations (a challenging requirement was that a temperature change of 0.2° in 7 seconds was specified at the interface surfaces). The optical system makes use of a modified Michelson interferometer while the mechanical system automatically compensates for thermal expansion effects. Originally the idea was to make a complete Zerodur structure to eliminate the thermal effects. However it appeared to be possible to use a titanium structure with certain elements made from invar and aluminium. No need to say that this reduced risk and cost of the instrument drastically.

For high accuracy alignment of optical components in optical instruments TNO TPD has developed dedicated, monolithic, flexure-based alignment mechanisms, which provide accuracies below 0.1 μm or 0.1 μrad as well as stabilities down to tens of picometers per few minutes after locking. High resolution, high stability alignment mechanisms consist of an adjustment mechanism and a locking device. Complex monolithic flexure-based mechanisms are designed to align specific degrees of freedom. They are realized by means of spark erosion. The benefits of these mechanisms are no play, no hysteresis, high stiffness, a simplified thermal design, easy to assemble. Using alignment mechanisms a passive system can be maintained. Locking after alignment is mandatory to guarantee sub-nanometer stability. However, a high accuracy alignment will be disturbed again due to drift during locking if the locking device is not properly designed. Several low-drift locking devices have been designed and developed. The dedicated alignment mechanisms presented here are based on: (a) the results of an internal ongoing research program on alignment and locking and (b) experience with mechanisms developed at TNO TPD for high precision optical instruments, which are used in e.g. a white light interferometer breadboard (Nulling) and an interferometer with picometer resolution for ESA’s future cornerstone missions “DARWIN” and “GAIA”.

Modern mirror optics often consists of a limited number of elements, in which many aberrations may be attacked by adjusting only one element in five degrees of freedom, i.e. all degrees of freedom except the rotation around the optical axis. When the adjustment has to be reusable and mass and stiffness are of importance, a hexapod mechanism is 'the mechanism of choice' for this function. This choice holds even though the hexapod controls six degrees of freedom, while control of only five degrees of freedom is required, simply because there was no known configuration that does only control the required five degrees of freedom while maintaining the superior mass and stiffness properties of the hexapod. In this paper a mechanical configuration is presented that offers a worthwhile alternative for the simultaneous adjustment of five degrees of freedom (one rotation constrained), in the sense that: 1. The rotation around one axis is constrained by the mechanical configuration, meaning that only the required five degrees of freedom have to be controlled. This means only five instead of six actuators are needed, which results in an increase in reliability. 2. The mass and stiffness of the mechanism are comparable with the hexapod. 3. From a mechanical and control point of view the configuration is less complex than the hexapod.

The two most important characteristics of mechanical structures used in space or astronomy are accuracy and stability. Much time, money and energy is invested in achieving this. However because the resources are limited it is important to realize that the cost of an instrument is mainly fixed during the design process. Three factors can influence the design to a great extend which are: creativity, design rules and analysis. The most important tool for analysis nowadays is the computer. This one has become so powerful that even large structures are no problem to model in detail in FEM. It is even tempting to spend much time and effort in optimizing structures with the computer. However the basis for the best result is creativity during the design phase and the application of design rules. Although design rules are used extensively it is surprising to see that one of the most important ones is so little used. This lecture is therefore especially about the rule to make structures static determined. The importance can be invaluable because in applying this rule there is no need for computers. Furthermore the application often results in non-conventional structures and it helps to clearly define the basics of the structure. It also provides simple qualitative results that help to make decisions concerning alternatives. Results of computer models can be verified on their validity. Some interesting results of the application of this design rule will be shown. Examples are the linear guiding system for the VLTI delay line and the mount of mirrors. Also existing structures like e.g. the secondary mirror mount of the VLT will be analysed and it will be shown how they could have looked like when they would have been static determined.

Opto-mechanical instruments are sensitive to temperature effects. The optical performance will be influenced by temperature variations within an instrument. Temperature variations can occur due to environmental or internal heat sources. Assembly at a different temperature than eventual operation of the instrument can also influence the performance. This paper describes principles to minimize thermal disturbance of optical performance. The thermal behaviour of a system can area-wise be divided in heat source, heat transfer area and place where the optical performance is affected. Placement of the heat source is critical. Using a large thermal capacity, the influence of the source will be minimized. Heat transfer can be controlled by insulation or by good conduction, the latter minimizing the thermal gradient along the thermal path. Thermo mechanical effects on the optical performance can be controlled using a thermal centre, a combination of materials with different expansion properties, low thermal expansion materials and scaling effects of the optical design. TNO TPD designs and manufactures opto-mechanical instruments for space and astronomy. The design guidelines described are commonly used in these instruments. Several examples of the application of these design guidelines are presented in this paper.

An optimal instrument control strategy for a dual-stage actuator system is presented. Resolution extension by signal conditioning and digital averaging techniques was implemented during the instrument design. The motion controller was developed based on robust PID (proportional-integral-derivative) algorithm and coarse/fine relay mechanism. The closed-loop residual noise and optimal dynamic performance are demonstrated by experiments.

This paper profiles the initial phase in the development of a six degree-of-freedom robot, with 1 μm dynamic positioning uncertainty, for the manipulation of x-ray detectors or test specimens at the Advanced Photon Source (APS). While revolute-joint robot manipulators exhibit a smaller footprint along with increased positioning flexibility compared to Cartesian manipulators, commercially available revolute-joint manipulators do not meet our size, positioning, or environmental specifications. Currently, a robot with 20 μm dynamic positioning uncertainty is functioning at the APS for cryogenic crystallography sample pick-and-place operation. Theoretical, computational and experimental procedures are being used to (1) identify and (2) simulate the dynamics of the present robot system using a multibody approach, including the mechanics and control architecture, and eventually to (3) design an improved version with a 1 μm dynamic positioning uncertainty. We expect that the preceding experimental and theoretical techniques will be useful design and analysis tools as multi-degree-of-freedom manipulators become more prevalent on synchrotron beamlines.

Two-Axis Rotation Systems, or “goniometers,” are used in diverse applications including telescope pointing, automotive headlamp testing, and display testing. There are three basic configurations in which a goniometer can be built depending on the orientation and order of the stages. Each configuration has a governing set of equations which convert motion between the system “native” coordinates to other base systems, such as direction cosines, optical field angles, or spherical-polar coordinates. In their simplest form, these equations neglect errors present in real systems. In this paper, a statistical treatment of error source propagation is developed which uses only tolerance data, such as can be obtained from the system mechanical drawings prior to fabrication. It is shown that certain error sources are fully correctable, partially correctable, or uncorrectable, depending upon the goniometer configuration and zeroing technique. The system error budget can be described by a root-sum-of-squares technique with weighting factors describing the sensitivity of each error source. This paper tabulates weighting factors at 67% (k=1) and 95% (k=2) confidence for various levels of maximum travel for each goniometer configuration. As a practical example, this paper works through an error budget used for the procurement of a system at Sandia National Laboratories.

The Geoscience Laser Altimetry System (GLAS) is a laser altimeter and LIDAR instrument on the Ice, Clouds, Environment Satellite (ICESat) mission. GLAS used 3 Nd:YAG lasers with 40 Hz rep rates at 4 Watts. All 3 lasers had to fire along a common beam path. Several mechanisms and optical assemblies were developed to allow the 3 lasers to fire down a common transmit path and exit the instrument. In the receive path of the GLAS instrument altimeter, there was a primary and redundant altimeter detector. A mechanism was designed, fabricated, and tested which would divert the incoming altimeter beam path from one detector to another. This mechanism was functionally similar to the mechanisms used on the transmit path. The Solar Ozone Limb Sounding Experiment II (SOLSE2) instrument had a requirement for rotating a visible (VIS) or ultra-violet (UV) filter into the instrument optical path. Both GLAS and SOLSE2 had similar operational and survival environments and lifetime requirements. A novel, precision rotational latching mechanism was designed to fulfill the requirements of both missions. The GLAS instrument had driving stability and repeatability requirements, such that if the mechanism met these stringent requirements, it would more than surpass the required performance for the SOLSE2 mechanism. The resulting mechanism, referred to as a “select mechanism” since it allows selection between 2 positions, was successfully designed and implemented for both missions. This paper describes the transmit path optical structures and select mechanisms of the GLAS & SOLSE2 instruments.

The compact and stable dual fiber optic refractive collimator is a device that takes laser light from two fiber optic cables, generating two beams of collimated light at increased diameters. This device is designed to be stable over a specific soak temperature range and maintain alignment through adverse vibration. Single or multiple beam configurations are possible with this design and the complexity goes linearly with beam quantity. What makes this device noteworthy is the simplicity of design plus ease in assembly and alignment. A precision alignment fixture is used instead of cumbersome, built-in, multiple degree of freedom features such as adjustment screws and flexures. This allows the collimator itself to be quite simple, compact and thermally stable. The dual collimator consists of a common housing, two fiber tip shuttle plugs allowing for adjustment in focus, and two laterally adjustable lens cells for beam alignment. The design has integral adhesive tack bonding features throughout and contains few parts keeping fabrication and alignment costs down. This paper covers the requirements, design, manufacture, assembly and performance of this optical device. The collimator has utility in precision interferometry. A patent has been filed.

Optical pointing and scanning mechanisms require inter-related optical, mechanical and electrical engineering and manufacturing disciplines. Such devices are employed in extremely diversified fields of photographic imaging, forward looking infrared (FLIR) systems, laser projection displays (LPD), target acquisition, aerial reconnaissance, remote sensing, and free space laser telecom. The degrees of freedom commonly applied include rotary scanning, raster and vector scanning, and limited angle gimbals. Support systems include flexures, ball bearings, and gas bearings. The performance of the optical payload supported by the pointing or scanning mechanism is paramount and dominates the process of materials selection, structural analysis, actuator selection, and control system development. This paper introduces the tradeoffs among range and type of motion, actuator types, angular sensor types, bearing types, and control systems applied to these types of systems. Actual product design and performance data is presented for a high-speed rotary scanner, a fast “nodding” scanner for a FLIR, a flexure supported fast steering mirror (FSM), and several ball bearing and a gas bearing gimbal designs.