Shack-Hartmann sensors are widely used to measure wavefront aberrations. We present the fundamental and specific engineering steps in the design of Shack-Hartmann wavefront sensors. Typical performance requirements such as sensor dynamic range, sensitivity and accuracy are defined and discussed. We investigate the trade-offs between these performance metrics and the factors affecting the trade-offs. A first order approach for selecting the optimal parameters of the sensor central piece, the lenslet array, is presented. We also propose a quick tolerance analysis method that can predict the wavefront measurement error due to misalignments, using only the ray-tracing software.

Field correcting lenses have been used to increase the fields of view of large telescopes since at least Ross's work in 1935. In recent years, due to the advent of electronic image sensors, there has been movement toward combining field correction with focal length reduction in so-called field compressor/corrector or telecompressor lenses. Such lenses are now commonly employed in both professional and amateur astronomy. This paper demonstrates that field compressor/corrector lenses can be of great utility in a wider context - the design of optical instruments. In the finite conjugate and relay applications prevalent in optical instrument design, standard achromats are often employed because there is currently no better alternative short of custom lenses. I show that there exists a particularly attractive family of cemented doublet compressor/correctors that can be combined with standard achromats to greatly improve the imaging performance obtainable in these applications. I propose that these field compressing/correcting doublets should be made available from stock as standard optical components. I introduce a novel viewpoint for analyzing the imaging capability of a lens that makes it simple to visualize the performance obtainable with a lens when it is employed over a wide range of fields and focal ratios. Using the new viewpoint, I demonstrate the performance improvement obtainable when employing the new compressor/correctors. I also show that the new lenses are flexible, that is, that one may obtain their advantages while employing them with a range of achromats or over a range of focal lengths, and that the alignment tolerances between the achromat and the field compressor/corrector are modest.

In this paper I will describe the design of a warm shield used in an infrared system that operates in the 8-12 micron wavelength region. I will describe the optical design considerations, show the experimental setup, describe the testing process and then I will evaluate the performance of the warm shield dewar against proposed requirements.

Optical lenses and systems have historically been of interest to the design and simulation community due to their extensive use in a variety of applications, including telecommunications and imaging. To simulate these systems, it is often possible to take advantage of symmetry or apply physical approximations, such as ray tracing and/or beam propagation, in order to avoid a rigorous solution of Maxwell's equations. However, in cases where these techniques are not applicable, such as when feature sizes become comparable to the wavelength of interest, full-wave electromagnetic simulations are required. Unfortunately, the time constraints and simulation sizes associated with many designs render this analysis infeasible or impractical, a problem that is aggravated by iterative design flows. Hence, a numerical platform capable of handling such computationally intense problems is necessary. In this paper, we present a hardware-based, full-wave simulation platform and demonstrate the advantages of hardware acceleration for the simulation of optical applications. Specifically, we will present a diffractive optical element (DOE) lens simulation, comparing results with previously published analytic and experimental data. We then modify the basic lens arrangement to produce an optical beam splitter. Because such a device cannot be analytically described using symmetry, this example will demonstrate the unique advantages of powerful hardware solvers for the analysis of DOE applications. Finally, we discuss how hardware-based solvers are critical for iterative designs by means of a 1-to-3 beam splitter example. This simulation system represents a new era in optical engineering, enabling the design of next-generation devices today.

In present-day optical system design, it is tacitly assumed that local minima are points in the merit function landscape without relationships between them. We will show however that there is a certain degree of order in the design landscape and that this order is best observed when we change the dimensionality of the optimization problem and when we consider not only local minima, but saddle points as well. We have developed earlier a computational method for detecting saddle points numerically, and a method, then applicable only in a special case, for constructing saddle points by adding lenses to systems that are local minima. The saddle point construction method will be generalized here and we will show how, by performing a succession of one-dimensional calculations, many local minima of a given global search can be systematically obtained from the set of local minima corresponding to systems with fewer lenses. As a simple example, the results of the Cooke triplet global search will be analyzed. In this case, the vast majority of the saddle points found by our saddle point detection software can in fact be obtained in a much simpler way by saddle point construction, starting from doublet local minima.

Projection systems comprising micromechanical scanning mirrors are a promising approach for information display of any kind. If combined with advanced laser diodes as light sources, ultra-compact projection heads can be realized, as it will be shown in this contribution. Besides the laser, key component of the system is a special MEMS device, a twodimensional resonant micro scanning mirror. The laser beam formed by collimator optics is directed onto the micro scanning mirror. Then, the reflected beam describes a highly complicated Lissajous figure on the projection screen with flare angles of up to 20 degrees. By driving the mirror and electrically modulating the intensity of the laser beam in a synchronous manner, projection of images can be achieved. Advanced techniques that guarantee improved image quality and allow compensation of artifacts because of relative movement between projection head, screen, and human observer will be described. Based on these principles, several optoelectronic systems have been designed. A monochrome projection head that incorporates the laser diode, optics and the micro mirror could be reduced to a volume of 15mm x 7 mm x 5mm. A slightly larger head is attached to a laser unit with red, green, and blue lasers via glass optical fiber for projection of full color images and video streams. All systems have VGA (640 x 480 pixels) resolution. They operate with 8 bit color depth per pixel and 50 frames per second. These key features in combination with the miniaturized size allow their use for a broad range of applications.

Aspherical glass or plastic lenses are usually adapted in the camera module of handheld phone. Recently, slimmer cameras are required according to reducing thickness of handheld phone. In this paper, we present an ultra slim camera module using multiple freeform off-axis imaging lenses. New optical concept of multiple lens system introduces the planar optics with freeform shaped aspherical lens surface on the wafer, which can achieve the thickness of 50% compared to the conventional symmetric lens system. In order to achieve the resolution specification, we separate the field of view in camera module. As the result, two inverted images are produced on a same imaging sensor and the acquired inverted images are processed by photo stitching algorithm so as to combine them. Finally, in order to verify the new imaging system, we manufacture the new slim lens through the UV embossing replication process. We found the new imaging system is feasible for VGA resolution and it can be expandable to the high resolution system.

Reliable fabrication and assembly of high-quality electro-optical imaging systems is a critical challenge facing electro-optical imaging system manufacturers. Optical compensation is one standard approach for minimizing the effects of errors introduced during the manufacture and construction of the optical subsystems. We describe how digital image processing should be considered as a form of compensation when evaluating a complete imaging system. We describe a novel method for digital-optical compensation which jointly adjusts both optical parameters and image processing parameters to maximize end-to-end imaging performance. We verify the superiority of this joint compensation strategy over the traditional sequential compensation through several example imaging systems.

NASA is endeavoring to launch missions capable of detecting Earth-like planets around neighboring stars. In visible wavelengths, this requires better than one 10 to the minus ten suppression of scattered light as close as 50 milli-arcsec to the stellar image. This extraordinary requirement is within reach but it requires broad-band wave front control to sub-Angstrom levels. We describe several high dynamic range imaging solutions, describe the various factors that contribute to the scattered light level and introduce a novel closed-loop broad-band correction system, suitable for the Shaped Pupil Coronagraph and the Lyot Coronagraph.

The goal of the Terrestrial Planet Finder Mission is to detect and characterize Earth-like planets. Detection of these faint objects, which appear very close to their parent stars, requires a coronagraph capable of achieving better than 10^-10 starlight suppression within a few Airy rings of the stellar image. The coronagraph is also required to maintain this high stellar extinction over a 100nm spectral bandwidth. To ease requirements on the telescope, a high planet light throughput and low sensitivity to wave front aberrations are also desirable features. An optical vortex coronagraph is a promising candidate architecture, which makes use of a spiral phase plate placed in an intermediate image plane to null out the stellar signal. This architecture has the advantage of high stellar extinction, high planet light throughput, and low sensitivity to wave front aberrations. Here we report the high contrast performance of an optical vortex coronagraph limited by the manufacturability of the spiral phase plate.

A comprehensive suite of Adaptive Optics systems and AO-assisted instruments is currently under development for the VLT and will be built around a hyperbolic convex adaptive Deformable Secondary Mirror (DSM). In telescopes with such a secondary mirror, test and calibration of both the DSM and the science instruments using it are notoriously expensive and time-consuming. The Adaptive Secondary Setup and Instrument Stimulator (ASSIST) is being developed to allow test and integration of three key elements of the future VLT Adaptive Telescope Facility: the DSM and 2 instrument-specific adaptive optics systems (GALACSI for MUSE and GRAAL for HAWK-I). The core of ASSIST is a standalone interferometric test setup for the DSM allowing its test at its center of curvature (significantly reducing the size of the classical hyperbole focii test configuration). This setup is completed by a star simulator for both natural and laser guide stars, a turbulence generator (for realistic AO performance measurements), a corrector system generating a VLT-like exit pupil and a Nasmyth rotator simulator interfacing with the two AO systems. In this paper we present the optical architecture of ASSIST, detailing the current design of the different parts of the system, and discuss its projected performance and compliance with the test and calibration requirements.

Recently, since micro optics have been widely used in many optical systems, it is desired to evaluate correctly the performance of such a micro optics and design them properly. In order to estimate accurately the diffraction effect, rigorous inclination factor of diffraction should be utilized. In this paper, we derive theoretically the self-consistent inclination factor which satisfies the reciprocity theorem in scalar imaging theory. By calculating numerically the point spread function in micro optics of no aberration, we also confirm the self-consistency of this factor. That is, the point spread function calculated by the diffraction at a spherical surface on the pupil coincides with that calculated by the diffraction at a plane on the same pupil. Our result will be very useful for evaluating and designing micro optics.

For the purpose of this paper, actual field curvature is the field dependent departure of the location of best image quality from a nominal surface along the direction of the optical axis. Usually this nominal surface is a plane where an image is formed and acquired. To the extent that location of best focus is displaced from it, the image is degraded due to defocus. Because the departure follows a curve in general, image quality over the field of view is compromised. Various image quality metrics can be used to calculate actual field curvature. These can also be used to generate contours of equivalent image quality, or isoquals, which are orthogonal to actual field curvature. This yields a solid method for evaluating how good the image quality of an optical system must be. It also yields a method for gaging geometric image quality versus diffraction and spawns new definitions for diffraction limited image quality.

We are witness to vehicle lamp technology experiencing significant development in the realm of halogen and high intensity discharge (HID) sources. As this is being written the ubiquitous light emitting diode (LED) is rapidly approaching headlamp reality. Government regulatory requirements challenge design, manufacture, and installation of headlamps. At the design stage lamp configurations use reflector and projector concepts. While reflector headlamps dominate the field, HID source small arc size often takes advantage of projector design by accomplishing both high and low beam settings using optical stops. These stops not only are a simple method of creating sharply defined beam tops, but also define beam top edges which aid significantly in aim adjustment. Evaluating quality of vertical beam alignment at vehicle assembly is often difficult to accomplish visually. While government regulations and automotive industry generic standards strive to define beam tops for ease of visual aim evaluation (audit), optical sensing offers improved audit repeatability and reliability. Therefore, new vehicle warranty claims related to headlamp beam vertical alignment can be minimized or even brought under complete control by proper vertical aiming and audit. An added result of well-managed headlamp beam alignment and audit at vehicle assembly is improved safety during night driving.

The problem of developing the lateral surfaces of a 3D object can arise in item inspection using automated imaging systems. In an industrial environment, these control systems typically work at high rate and they have to assure a reliable inspection of the single item. For compactness requirements it is not convenient to utilise three or four CCD cameras to control all the lateral surfaces of an object. Moreover it is impossible to mount optical components near the object if it is placed on a conveyor belt. The paper presents a system that integrates on a single CCD picture the images of both the frontal surface and the lateral surface of an object. It consists of a freeform lens mounted in front of a CCD camera with a commercial lens. The aim is to have a good magnification of the lateral surface, maintaining a low aberration level for exploiting the pictures in an image processing software. The freeform lens, made in plastics, redirects the light coming from the object to the camera lens. The final result is to obtain on the CCD: - the frontal and lateral surface images, with a selected magnification (even with two different values for the two images); - a gap between these two images, so an automatic method to analyse the images can be easily applied. A simple method to design the freeform lens is illustrated. The procedure also allows to obtain the imaging system modifying a current inspection system reducing the cost.

Changes in the shape of large lens elements due to the influences of gravity are important to consider in the fabrication, testing and assembly of optical systems. Tried and proven methods used for mounting large mirrors to minimize the effects of gravity are typically not applicable to large transmissive lens elements, due to the simple requirement that the clear aperture of a lens must remain free of mechanical obstructions. Precautions must be taken to ensure that an element's surfaces are correctly fabricated and then maintained when assembled into the final system. The amount of distortion caused by the weight of a particular lens element is dependent on a number of factors including: size, aspect ratio, shape, material, and the support on which it rests. Examples of the effects of these factors are modeled using Finite Element Analysis and demonstrated through interferometric testing. Attention is given to the mounting of lens elements within a system and simulating "real-world" conditions. These "real-world" conditions can produce results that are different from what was expected if only ideal cases have been considered. The work presented will aid the designer, fabricator, and metrologist to identify what optical elements and mounting conditions may be problematic and to minimize their effects.

The optomechanical engineering for mounting lenses and mirrors in imaging systems is frequently driven by the pointing or jitter requirements for the system. A simple set of rules was developed that allow the engineer to quickly determine the coupling between motion of an optical element and a change in the system line of sight. Examples are shown for cases of lenses, mirrors, and optical subsystems. The derivation of the stationary point for rotation is also provided. Small rotation of the system about this point does not cause image motion.

Deriving the optimal bondline thickness for an athermal bondline depends on many factors. An optimum bondline is defined as one that produces zero radial stress at the optic/adhesive interface. A review of the current equations in use and a new non-linear equation defined for optic mounts over larger temperature ranges is included. An assessment of sensitivities around the optimum bondline thickness is discussed. Also, guidelines for do's and don'ts in athermal optics mounts is included.

Chemical Mechanical Polishing, also referred to Chemical Mechanical Planarization (CMP), is one of the enabling technologies which allows the fabrication of high performance multi-level metal structures in IC fabrication. In this paper we will discuss the specific application of CMP techniques to aluminum mirror polishing and the resultant super polished finish obtained. Current aluminum mirror processing methods use combinations of machining, lapping and diamond turning operations to achieve required surface accuracy and quality. Optimum results from diamond turning yields surface figure with an error of no less than half a wave and surface roughness no less than 50 angstrom aluminum substrates. In addition, diamond turning puts "grooves" onto the surface that act as a diffractive element resulting in specular beam power loss and ghost images. Often these diffractive and scatter effects, inherent to grooved surfaces, are too severe to provide adequate performance in the UV and visible range. Further, the low signal to noise ratio of the optical system reduces resolution and the overall efficiency of the optical system. A new procedure for polishing bare 6061-T6 Aluminum monolithic mirrors using Chemical Mechanical Planarization (CMP) slurry and techniques yields extremely high quality, low scatter mirrors. Planar aluminum mirrors with flatness equivalent to lambda/10 and R_a <2 nm have been polished and measured on a Veeco NT3300 white light Interferometer (at 20X). Comparison of the power spectral density curves of mirrors produced via CMP with those presently produced with diamond turning shows reduction across the range of spatial frequencies (1-10³ mm^-1) and elimination of the grooving frequency. Both white light interferometer and AFM images show the polished surfaces to be smooth, pit free with no pull outs.

The ZERODUR production, consisting of established processes used in the manufacturing of high homogeneous optical glasses, results in excellent blanks with low stress birefringence, striae content and outstanding homogeneity of the coefficient of thermal expansion. For future extremely large telescope projects like OWL (OverWhelmingly Large Telescope) or TMT (Thirty Meter Telescope), with at least several hundreds of mirror blanks, the material homogeneity within a single blank is extremely important. Previously, the stress birefringence of 2m class mirror blanks could be reduced to amounts far below our catalog values. Striae in ZERODUR, if present, are weak band-like density fluctuations within the material with only minimum influence on the properties of the material. This paper presents the results of dilatometric measurements on the influence of standard grade striae within ZERODUR on the homogeneity of the CTE. All CTE measurements have been carried out using our new dilatometer setup with improved reproducibility.

CCOS (Computer Control Optical Surfacing) technology is widely used for making aspheric mirrors. For most manufacturers, dwell time algorithm is usually employed to determine the route and dwell time of the small tools to converge the errors. In this article, a novel damp iterative algorithm is proposed. We chose revolutions of the small tool instead of dwell time to determine fabrication stratagem. By using resistance iterative algorithm, we can solve these revolutions. Several mirrors have been manufactured by this method, all of them have fulfilled the demand of the designers, a 1m aspheric mirror was finished within 3 months.

Conventional microlens arrays consist of a repetitive arrangement of a unit cell on a fixed pitch. In a chirped array, the inflexibility of a regular structure has been overcome. Here, the array consists of individually shaped lenses which are defined by a parametric description of the cells optical function. We propose different fabrication methods for chirped microlens arrays and present experimentally obtained data. Reflow of photoresist is an established technology for the fabrication of microlenses with superior optical performance. For the generation of a chirped microlens array the photolithographic mask for patterning the resist to be melted has to be chirped. We present an algorithm for mask generation with an example of an ultra-thin camera objective. Inherent to the reflow process stringent limitations to viable surfaces apply. For achieving more arbitrary surfaces, laser lithography and also 2-photon polymerization are employed. In both methods the structures are decomposed into pixels. In laser lithography the local height is converted into an intensity value for the exposure. This variable dose writing locally changes the solubility of the resist in the development process leading to the required surface profile. We propose a writing scheme enabling structure heights of several ten microns with sufficient height discretization. 2-photon polymerization is a rapid prototyping method. Here, a small volume of a UV-curing organic-inorganic co-polymer is hardened in the tight focus of the writing beam. The volume pixel to be exposed is addressed by piezoelectric translation stages. Experimentally obtained structures and performed tests of the optical quality are presented.

This paper reports on the commissioning of the first of Zeeko's "IRP1200" 1.2m capacity 7-axis automated CNC polishing machines. These combo machines now support five different removal regimes, which are described. The machines differ substantially from Zeeko's more familiar 200mm machines on which we have focused before, in terms of overall architecture and detailed design. Large and small optics place different demands on part-fixturing, tooling, machine speeds and accelerations, metrology, slurry-handling, part-loading and access etc. These have had a profound effect on the development-path from 200 to 1.2m machines. Moreover, an advance in the kinematic design has extended the allowable range of surface slope-angles from typically 30° up to a hemisphere. The paper presents results from the pass-off trials, the first fluid-jet experiment, and the development of tooling to address a requirement to smooth a part with a local defect.

Magnetorheological finishing (MRF) is a computer controlled polishing process (CCP), which is commonly used in the field of high quality optical lens production. The process uses the material removal characteristic of the polishing tool (influence function) and the surface error-profile to calculate individual, surface error-profile dependent polishing sequences. At the University of Applied Sciences Deggendorf a testing series with a magnetorheological finishing machine has been performed, and effects of the influence function size and its removal capacity on the polishing quality and the process time have been investigated. The result of the research shows that the influence function size has a major effect on the process time, whereas the polishing quality is nearly independent of the influence function size. During the testing series the process time was significantly reduced using an appropriate influence function size. The process time decreased about 9% relating to the original influence function.

Magnetorheological finishing (MRF) is a computer controlled polishing (CCP) technique for high precision surfaces. The process uses a magnetorheological fluid which stiffens in a magnetic field and thus acts as the polishing tool. A standard MR fluid consists of magnetic carbonyl iron (CI) particles, nonmagnetic polishing abrasives and liquid. To delaying oxidation of the iron particles and avoiding agglomeration the liquid consists of water completed with stabilizers. For the material removal and smoothing of the surface mostly cerium oxide or diamond is used. The materials to be polished may tend toward to different sedimentations of the MR fluid on the machined surface. These sedimentations result from the machining and may develop a polishing layer with MR fluid components. At the University of Applied Sciences Deggendorf analysis of the machined surface are made by the scanning electronic microscope (SEM) to determine the sedimentations of the finishing. The results of the research display the influence for the surface properties due to developing polishing layer by magnetorheological finishing.

This study presents the design and analysis of a lens with a variable focal length, which we will call an adaptative lens. This lens is formed of two transparent elastic surfaces with a transparent liquid medium between them. The mechanical design of the lens considers the variation of liquid pressure between the surfaces. This causes changes in curvature radii and in the axial thickness, generating variations in the focal length of the lens. An analysis is given of changes in lens power with respect to changes in pressure. Finally, a model of the human eye is presented using this adaptative lens as crystalline.

This study presents results obtained from the exact tracing of rays of an adaptive lens, that is, a liquid lens with transparent elastic surfaces. Because the elastics surfaces are deformed by a liquid, they acquire different curvatures depending on the difference in pressures of the liquid and of the environment. Images generated by this type of lens are simulated with a ray-tracing computer program, considering that each surface can be modeled with concentric spherical rings with different curvatures; this is because the elastic coefficients of the plastic materials used are non-linear.

Most microscope objectives commonly used for two-photon imaging of neurons are optimized for high image quality under wide-field illumination and for long working distance, constraints that are at odds with the need for high fluorescence collection efficiency and large field of view dictated by the low fluorescence intensity and the scattering properties of brain tissue. We present a de novo design of an objective intended specifically for deep-tissue functional two-photon imaging. Our design has separate imaging and non-imaging pathways for incident and emitted light, making use of the optical sectioning intrinsic in non-linear fluorescence excitation, which relaxes a number of design constraints. We show through modeling that a twofold to fourfold improvement in fluorescence collection efficiency over traditional objective designs is easily achievable while maintaining nearly diffraction-limited performance within a 200-micron field of view and a long working distance.

The characterization of the residual birefringence of erbium-doped fibers using the Poincare sphere requires a careful alignment of the polarimetric set up. It is necessary to define the reference frame and then, to measure the input and output polarization states. The procedure is time consuming when the sample and the prism polarizer are removed and repositioned manually. In this work we present an automatic polarimetric set up that takes advantage of the geometric properties of the Poincare sphere and the polarization optics of homogeneous retarders. The measurement procedure has been modified to improve the measurement precision and to shorten the time required to perform the spectral birefringence characterization of these laser fibers. We discuss the fundamental limitations of this evaluation procedure comparing it with the wavelength scanning method based on Jones calculus, often used to characterize the birefringence of optical fibers.

This paper investigates the qualified method of the removal of hot spot speckle produced by a green laser with long coherence length for a projection display system. The application of a random phase plate using a binary computer generated hologram pattern is successfully applied to reduce the speckle contrast ratio to 2%; this is below the minimum level of 4% which can be recognized by the human eye.

Basing on the principle of dynamic active confocal measuring method, a non-contact probe was introduced. The relation between the displacement of the object measured and the displacement of the oscillating unit was established by applying to the measure principle. To find the effect way to improve the output signal, a mathematical modal of the photoelectric signal was proposed. According to the mathematical modal, the diameter of the pinhole was decided and the optical path was adjusted, which improved the performance of the probe. It valuable to emphasize that the double lens oscillating unit was the crucial parts of the mechanical and optical structure. The tuning fork stimulating, inductive coils and the pinhole and photodiode adjuster device were designed according to the theoretical analysis, which were more convenient to the measure task. The stability of the oscillating unit was tested; and the optical and mechanical performance of the probe was validated by experiments using gage blocks combining with a micro-motion measuring mount model. The cost of a micro-displacement testing sensor was saved in designing of this kind of probe. A more accurate signal processing method was discussed in this paper.

Cryogenic optical systems are often mounted in vacuum cryostat. An interesting concern in the mechanical design is how to maintain the stability of the line of sight (LOS) of the cryogenic optical system when it is cooled down to very low temperature. To investigate this problem an experimental cryogenic optical system, its mounting structure, and a cryostat were designed and fabricated. After assembling them together, a test experiment for measuring the LOS drift was carried out. Finally, the test results will be presented in this paper.

Thermal imaging is an important, though challenging, diagnostic for shockwave experiments. Shock-compressed materials undergo transient temperature changes that cannot be recorded with standard (greater than ms response time) infrared detectors. A further complication arises when optical elements near the experiment are destroyed. We have designed a thermal-imaging system for studying shock temperatures produced inside a gas gun at Sandia National Laboratories. Inexpensive, diamond-turned, parabolic mirrors relay an image of the shocked target to the exterior of the gas gun chamber through a sapphire vacuum port. The 3000-5000-nm portion of this image is directed to an infrared camera which acquires a snapshot of the target with a minimum exposure time of 150 ns. A special mask is inserted at the last intermediate image plane, to provide dynamic thermal background recording during the event. Other wavelength bands of this image are split into high-speed detectors operating at 900-1700 nm and at 1700-3000 nm, for time-resolved pyrometry measurements. This system incorporates 90-degree, off-axis parabolic mirrors, which can collect low f/# light over a broad spectral range, for high-speed imaging. Matched mirror pairs must be used so that aberrations cancel. To eliminate image plane tilt, proper tip-to-tip orientation of the parabolic mirrors is required. If one parabolic mirror is rotated 180 degrees about the optical axis connecting the pair of parabolic mirrors, the resulting image is tilted by 60 degrees. Different focal-length mirrors cannot be used to magnify the image without substantially sacrificing image quality. This paper analyzes performance and aberrations of this imaging diagnostic.

The size and the focal length of camera objectives (e.g. cell phones or digital cameras) are becoming smaller and smaller. At the same time the quality requirements are increasing. Besides surface accuracy, the imaging quality of the complete optics is mainly influenced by the alignment errors of the single elements. TRIOPTICS has developed a new technology called MultiLens in order to measure the centering errors of all single surfaces within an objective lens with up to 40 surfaces or more. We achieve accuracies in the range of an arc second. During the measurement the deviation of each center or curvature with respect to a reference axis is measured. These data are further processed in order to provide the shift and tilt of an individual lens or group of lenses in respect to a given reference axis (Patent pending Ref. 1). Applications mainly include the measurement of cell phone and digital camera lenses. However, any type of objective lens from endoscope up to very complex objective lenses used in microlithography can be measured with highest accuracy. The method has been extended to measure also the aspherical axis of lenses.