We describe the optical fabrication and the active support system of the 6.5 m f/1.25 primary mirror for the first Magellan telescope. Figuring was performed with a 1.2 m stressed lap, which bends under active control to match the local curvature of the optical surface, and small passive tools. The figure was measured with IR and visible interferometers, using refractive null lenses to compensate 810 microns of aspheric departure. After subtraction of Seidel astigmatism and spherical aberration, the finished mirror is accurate to 14 nm rms surface and has an encircled energy of 80% in 0.06' diameter at 500 nm. The mirror was integrated with its active support system in the laboratory, and support forces were adjusted to optimize the figure. The optimization was performed by singular value decomposition of the influence functions into normal bending modes. Using the first 20 modes and a maximum correction force of 46 N, the surface accuracy is 24 nm rms with 80% of the light in 0.11' diameter.

In 1989 and 1994, REOSC was awarded contracts for the optical fabrication of the four VLT and two Gemini 8-m class primary mirrors, respectively. We elaborate on the mirror concepts and their optical specifications, and discuss the fabrication and testing methods applied to the production of these mirrors. We also provide an overview of the facilities and equipment designed and built to transport, handle and process the mirrors. All units have been completed well within requirements; we provide detailed insight into the results obtained with all mirrors, and demonstrate that very large, aspheric mirrors can be produced to utmost accuracy-- in practice, well below the diffraction limit.

The Gran Telescopio Canarias (GTC) is a 10 m-class telescope which is under construction and will be operational at the Observatorio del Roque de Los Muchachos at the end of 2003. The goal of this paper is to describe the current status of the design and construction of the primary, secondary and tertiary mirrors of the GTC and their opto-mechanical supports. It also summarizes the optical performances expected from the GTC and the error budget of the optical system.

The elastic methods, especially the large deformations theory, and their applications to optical manufacturing are discussed in the frame of the next telescope generation. The force distribution corresponding to the required aspheric deformation can be applied during the figuring and polishing process or later on, directly in the telescope. For highly aspheric optics, like those considered in optical designs for the OWL project, the first solution, also called stress polishing, is preferred. Calculation and results obtained for a 6 - 8 m class mirror with the high asphericity foreseen in the OWL designs are presented.

The primary mirror of the proposed California Extremely Large Telescope is a 30-meter diameter mosaic of hexagonal segments. An initial design calls for about a thousand segments with a hexagon side length of 0.5 meters, a primary-mirror focal ratio of 1.5, and a segment surface quality of about 20 nanometers rms. We describe concepts for fabricating these segments.

On the basis of its experience gained on the VLT and GEMINI primary mirrors polishing REOSC will present his first comments on the preliminary technical specifications of some concepts (OWL, MAXAT) discussed with their instigators (ESO, AURA). Similarities with other projects presently running (CEA LIL) at REOSC that lead to consider such type of giant telescope with some unsuspected serenity. In addition the discussion will be illustrated with the REOSC experience in medium and large size off axis aspheric optics fabrication, the most recent being the Gran Telescopio Canaries project for which REOSC has just been awarded by GRANTECAN S.A. for polishing the 36 segments of this 11-m instrument.

Two years after the first light of UT1 a summary of the On- site optical maintenance situation seems necessary. A comparison with the Maintenance Program developed in Europe will be commented. Measured data on the large mirrors and the efforts to maintain and improve the on-site contamination at a low level will be presented. Data and pictures will illustrate the CO2 in-situ cleaning technique, advantage and disadvantage. Methods and results of the AMOS washing unit, in the preparation of mirrors before coating, will be presented. Emphasis will be put on the reflective Al coating quality of the LINDE coating unit and the achieved reflectivity data demonstrated.

Ceramic mirrors are going to become more and more attractive for realizing lightweight optomechanical systems. The C/SiC ceramic described in this article, a joint development of DSS and IABG offers design freedom through a novel straight forward manufacturing process, combined with excellent optomechanical properties. This process not only enables to build ultra lightweight mirrors of very high complexity, but also very large 3d-structures. Mirrors up to 80 cm and structures up to 2.5 m have been realized. The paper will summarize the recent progress achieved in producing mirror blanks up to 1 m diameter and beyond. Recent results in new technologies for achieving high performance optical surfaces will be reported. An outlook for future envisaged applications of C/SiC technology, e.g. 2.5 m mirror segments for the next generation of large earth and space based telescopes, is given.

This paper will discuss the state of these technologies: Vacuum Hot Press Bonding, Near Net Shape Forming with re- useable mandrels, and the development of the O-30 material. The paper will also attempt to show how these new manufacturing technologies addresses the perception that beryllium is an expensive, long lead material.

A CNC-controlled precision measuring machine is a very powerful tool in the optical shop not only to determine the surface figure, but also to qualify the radius of curvature and conic constant of aspherics. We used a commercially available 3D-coordinate measuring machine (CMM, ZEISS UPMC 850 CARAT S-ACC) to measure the shape of the GEMINI 1-m convex secondary mirrors at different lapping and polishing stages. To determine the measuring accuracy we compared the mechanical measurements with the results achieved by means of an interferometrical test setup. The data obtained in an early stage of polishing were evaluated in Zernike polynomials which show a very good agreement. The deviation concerning long wave rotational symmetrical errors was 20 nm rms, whereas the accuracy measuring of mid spatial frequency deviations was limited to about 100 nm rms.

An interferometric test is described that allows the testing of a convex surface from its reflective side with conjugate off-axis light beams. The apertures of the auxiliary optics are smaller than that of the surface under test. A method of centering the convex surface in the testing beams is presented.

The Tertiary Mirrors for the ESO Very Large Telescope project consist of four optical flats (elliptical, 890 X 1260 mm²). The achieved opto-mechanical design is challenging since it provides high optical overall quality combined with high stiffness (70 Hz Eigenfrequency) and low mass (total mass of 180 kg for the complete unit). Schott (Mainz, Germany) produces the lightweight Zerodur blanks. Carl Zeiss has designed and manufactured the mirror and its support cell. Last not least it became necessary to install the biggest testing equipment for flats in Europe to guarantee for a scientifically correct verification of the quality of the complete unit. All four mirrors have been delivered to ESO.

In a recent paper McLeod proposed and used measurements of the field dependence of third order astigmatism to collimate a Ritchey-Chretien telescope with the stop at the primary mirror. We adapt this method to the Cassegrain focus of the ESO Very Large Telescope, where the stop is at the secondary mirror and the telescope is only corrected for spherical aberration. In addition, we study the effects of the practical definition that the center of the field is the center of the adapter. We present measurements of the field astigmatism and discuss the accuracy of this collimation method.

The Galileo Telescope is a Ritchey-Chretienne telescope with an aperture of 3.58 meters and is located in La Palma Island. Since the first light in June 1998, several nights have been used to calibrate and to test the active optics systems, an operation involving also the control and reduction software and the optics of the de-rotator system. At the moment the active optics system routinely works at the beginning of the nights to support both technical and scientific observations. The ultimate tests on the control system are here reported and its performances are analyzed in order to quantify the final optical quality. Also a brief report on the effects of the serrourier temperature is reported. A comparison between Shack-Hartmann analysis performed on the rotator adapter bench and wavefront analysis on the instrument focal plane (at Nasmith station) is also reported. Some images taken in the visible and near infrared ranges during the same night are shown.

The SOAR Telescope project has embarked on the development of a very high quality 4.2-meter diameter optical telescope to be sited on Cerro Pachon in Chile. The telescope will feature an image quality of 0.18 arc seconds, a moderate field of 11 arc minutes, a very large instrument payload capacity for as many as 9 hot instruments, and an Active Optical System optimized for the optical to near IR wavelengths. The active optical system features a 10 cm thick ULE^TM primary mirror supported by 120 electro- mechanical actuators for a highly correctable surface. the 0.6 meter diameter secondary is articulated by a hexapod for real time optical alignment. The 0.6-meter class tertiary will provide fast beam steering to compensate for atmospheric turbulence at 50 hertz and a turret for directing the light to either of two nasmyth or three-bent cassegrain ports. Both the secondary and tertiary are light- weighted by machining to achieve cost-effective low weight mirrors. This paper discusses the unique features of this development effort including many commercial products and software programs that enable its technical feasibility and high cost efficiency.

We have recently commissioned an active-optics upgrade for Kitt Peak National Observatory's 4 m Mayall telescope. The active-optics upgrade project is based largely on the CTIO 4 m upgrade and consists of three principle subsystems: (1) the 4 m Active Primary Support, (2) secondary mirror articulation and (3) a dedicated wavefront sensing system.

Modern Acquisition and Guiding systems provide a range of services with the aim of optimizing telescope performance. These include wavefront sensing for active optics, fast guiding and measurement of seeing. On a segmented-mirror telescope, the Acquisition and Guiding system may also be expected to provide measurements of the piston errors between the primary mirror segments. The Guacamole (GUiding, Acquisition and CAlibration MOduLE) system of the 10 m Gran Telescopio Canarias (GTC) has been designed to provide all of these services. Complete systems will be installed at each of the GTC Nasmyth foci. The requirements of this system are presented and a preliminary design is described.

The active optics system of the ESO Very Large Telescope has now been in operation since May 1998. All results from the wavefront analyses as well as numerous other telescope and environmental data are continuously logged. This allows for an accurate assessment of the performance of the active optics system and yields statistical correlations between the wavefront data and other telescope or external parameters.

The Gemini primary mirror support incorporates a system of hydraulic whiffletrees to carry the mirror weight and define its position. The six orthogonal kinematic degrees of freedom are controlled by six hydraulic zones--three axial, two lateral, plus a transverse lateral. By varying the fluid volumes in these hydraulic zones the mirror position can be adjusted in all six degrees of freedom. Because of the finite lengths of the linkages that connect the mirror to the lateral supports, any shift in mirror position changes the amplitudes and directions of the applied forces with a resulting effect on the static balance and mirror figure. These effects have been calculated for mirror translations and rotations in all six degrees of freedom, resulting in predictions of the changes in the axial and lateral support forces and in the mirror figure. This paper describes the modeling as well as experimental verification of the results.

A sensing and control system for maintaining optical alignment of ninety-one 1-meter mirror segments forming the Hobby-Eberly Telescope (HET) primary mirror array is now under development. The Segment Alignment Maintenance System (SAMS) is designed to sense relative shear motion between each segment edge pair and calculated individual segment tip, tilt, and piston position errors. Error information is sent to the HET primary mirror control system, which corrects the physical position of each segment as often as once per minute. Development of SAMS is required to meet optical images quality specifications for the telescope. Segment misalignment over time is though to be due to thermal inhomogeneity within the steel mirror support truss. Challenging problems of sensor resolution, dynamic range, mechanical mounting, calibration, stability, robust algorithm development, and system integration must be overcome to achieve a successful operational solution.

We have developed and tested extensively three different methods for phasing the primary mirror segments of the Keck telescopes. Two of these, referred to as the broadband and narrowband algorithms respectively, are physical optics generalizations of the Shack-Hartmann technique. The third, Phase Discontinuity Sensing, is a physical optics generalization of curvature sensing. We evaluate and compare experimental results with these techniques with regard to capture range (as large as 30 micrometers ), run-to-run variation (as small as 6 nm), execution time (as short as twenty minutes), systematic errors, ease of implementation, and other factors, in the context of the Keck telescopes and also of future very large ground-based telescopes.

California Institute of Technology and University of California have begun conceptual design studies for a new telescope for astronomical research at visible and infrared wavelengths. The California Extremely Large Telescope (CELT) is currently envisioned as a filled-aperture, steerable, segmented telescope of approximately 30 m diameter. The key to satisfying many of the science goals of this observatory is the availability of diffraction-limited wavefront control. We describe potential observing modes of CELT, including a discussion of the several major outstanding AO system architectural design issues to be resolved prior to the initiation of the detailed design of the adaptive optics capability.

The primary mirror of the proposed California Extremely Large Telescope is a 30-meter diameter mosaic of hexagonal segments. The primary mirror active control will be achieved using four systems: sensors, actuators, processor, and alignment camera. We describe here the basic requirements of sensors and actuators, sketch a sensor design, and indicate interesting actuator alternatives.

Efficient image-based procedures have been developed for both tilt and piston alignment of the mirror elements of large segmented telescopes. Details of the algorithms and their implementation are described, and results are presented for both computer simulations and for laboratory experiments. The techniques are especially intended for space-based systems, but means are discussed for applying them to ground-based telescopes as well.

The results of a number of research projects related to the phasing of segmented telescope primaries are presented. The behavior of a segmented mirror controlled using edge position sensors is examined using the results of a numerical simulation. The performance of a novel approach to the optical sensing of piston differences is analyzed. The effect of segmented manufacturing errors on both the phasing control system and telescope performances is discussed. Finally, a concept for an `autonomous segment mirror' is presented and its feasibility assessed.

The model of distortions caused by mistakes in installation of mirrors of a compound telescope of the large diameter is offered. The results, received on the basis of model, calculated according to the formulas for average image in a plane of its formation, it variance and characteristic of speckle structure of the image are resulted. It is shown, that the researched distortions should be taken into account at a choice of an installation site of a telescope, at processing the information in a service of meteoric and asteroid safety and fundamental astrophysical researches.

The PSF of a segmented mirror telescope is hardly affected by the segments alignment and it can cancel the performances of the Adaptive Optics Systems. The piston and tilt of each segment must be uniformly adjusted in relation to the rest of the segments. Furthermore, the direct detection of the alignment error with natural stars would be desirable in order to monitor the errors during the astronomical observation. We have studied the lost information of the piston error by the presence of atmospheric turbulence in the measurements of curvature and present a new algorithm to obtain the local piston using the curvature sensor. A phase wrapping effect is shown as responsible for the loss of curvature information to get the piston error map well enough; this happens not only in the presence of atmospheric turbulence, but also without it.

Two characteristics of next generation extremely large ground-based telescopes are certain: (1) they will be segmented, with active control of segment position, tilt and, possibly, shape, and (2) they will have adaptive correction of atmospheric wavefront errors. Both these aspects place new constraints on wavefront sensing and wavefront correction. For quantitative analysis of the wavefront sensing requirements of these telescopes, it is necessary to analyze the effects of segmentation on image quality. To achieve this we are building a telescope simulator for calculating PSF and related image quality parameters: Strehl ratio, encircled energy, modulation transfer function, etc. A large number of segments (up to 10,000) may be simulated, each affected by piston and tilt errors. Piston and tilt errors are randomly generated to create realistic mirror configurations. Due to the large number of segments, interesting results are observed, both with respect to numerical statistics and diffraction. Strehl ratio as a function of rms piston error is found to depend upon the probability density function of the piston error, and the dominant effect in the case of segment tilt is that of diffraction from a 2D grating rather than the geometrically expected multiple image. Fine analysis of simulated point-spread function allows a discussion of the pertinence of different image quality parameters for the case of segmentation errors.

We demonstrate that the curvature equation can be modified using some properties of Distributions theory for segmented mirror techniques. It is shown that, additionally to the individual segment aberrations, the modified equation contains the information about the relative pistons and tip- tilts among the segments. The validity of the equation is verified by numerical simulations and by a laboratory experiment as well.

The design, parameters and the amplitude-frequency analysis of the new magnetic rheology (MR) drive are presented. The combination of hydrostatic carrier, MR hydraulic loop control, elastic thin wall seal joined in a single unit ensures small positioning error nm and small time of response T <EQ 200 ms.

Astronomical research requiring high photometric dynamic range is restricted by all current telescopes. Fundamental diffraction limitations and practical difficulties associated with conventional `centered' telescope optics and their mounts leads to unavoidable light scatter which can be the limiting noise source in observations of faint or diffuse objects in the presence of much brighter contaminating sources. We describe here an off-axis design for a 6.5 m astronomical telescope optimized for low scattered light and low emissivity. This is part of a new concept for a facility which we call the New Planetary Telescope (NPT). We shown how its geometrical optical performance can equal that of an on-axis conventional telescope while the diffractive performance fundamentally surpasses conventional telescopes because of the absence of pupil obstruction. The decentered NPT concept also allows wide-field and versatile instrumentation configurations that are not possible with more conventional designs.

Mercury liquid mirror telescopes work. This ia by now a well-established fact. However they suffer from a major limitation: they cannot be tilted. We have recently proposed that liquid mirrors can be tilted by several tens of degrees, provided one can develop a high-viscosity liquid having a high coefficient of reflectivity. The technology promises inexpensive large telescopes. We present the concept and some of the ongoing work.

The Sydney University Multiple Mirror Telescope (SUMMIT) is a medium-sized telescope designed specifically for high resolution stellar spectroscopy. Throughout the design emphasis has been placed on high efficiency at low cost. The telescope consists of four 0.46 m diameter mirrors mounted on a single welded steel frame. Specially designed mirror cells support and point each mirror, allowing accurate positioning of the images on optical fibers located at the foci of the mirrors. Four fibers convey the light to the future location of a high resolution spectrograph away from the telescope in a stable environment. An overview of the commissioning of the telescope is presented, including the guidance and automatic mirror alignment and focussing systems. SUMMIT is located alongside the Sydney University Stellar Interferometer at the Paul Wild Observatory, near Narrabri, Northern New South Wales.

This paper presents the recent progress in the optical system of LAMOST project: (1) The image quality at the edge of the field of view and the worst sky area is about 1.76 arc seconds. To improve this value, many aspherical lenses with 3 mm in diameter will be attached t some of 4000 optical fibers. With maximum 6.6 microns asphericity, these lenses could compensate the residual off-axis aberration of the optical system and obtain much better image quality, thus in any case the image quality could less than 0.6 arc seconds in diameter. (2) At the open air, an outdoor experiment program for creating the aspherical surface and accurately measuring the wave front shape of the active Schmidt correcting plate, which is with the full scale of the sub-system of the telescope, is introduced here. (3) Some further consideration and modification for the active support of each sub-mirror of the Schmidt correcting plate is discussed also in this paper. The presentation in this paper showed that the LAMOST is very likely to get much better quality even it is with a very large field of view, and its key technology--active optics is also reaching the realization.

SALT is a 10-m class telescope for optical/infrared astronomy to be sited at Sutherland, the observing state of the South African Astronomical Observatory. This telescope will be based on the principle of the Hobby-Eberly Telescope (HET) at McDonald Observatory, Texas. This cost-effective design is a tilted-Arecibo concept with a segmented spherical primary mirror of diameter 11 meters. The telescope has a fixed gravity vector but with full 360 degrees rotation in azimuth. A spherical aberration corrector mounted on a tracker beam at the prime focus enables a celestial object to be followed for 12 degrees across the sky. The SALT design enables over 70% of the sky to be accessed for about 20% of the cost of a conventional telescope of similar aperture. The telescope will be used primarily for spectroscopic studies of celestial objects with a light-weight low-dispersion imaging spectrograph mounted at the prime focus and other higher-dispersion instruments fiber-fed and mounted in an environmentally controlled basement. The concept design for SALT is presented with emphasis on the design changes between SALT and HET.

The Southern African Large Telescope (SALT) is a 10-m class telescope for optical/infrared astronomy to be sited at Sutherland, the observing station of the South African Astronomical Observatory. This telescope will be based on the principle of the Hobby Eberly Telescope (HET) at McDonald Observatory, Texas: a cost-effective design involving a tilted-Arecibo concept with a segmented spherical primary of diameter 11 meters. A spherical aberration corrector mounted on a tracker beam at the prime focus enables a celestial object to be followed for twelve degrees across the sky. This paper will discuss the correction of the spherical aberration for SALT, including presenting optical designs similar to the HET corrector, but with substantially better performance expected. Although the discussion will concentrate on the specific case of SALT, the scope is wide enough to be of interest for spherical aberration correction of similar telescopes (spherical primary feeding a tracking prime focus) which may be amongst the next generation of very large aperture instruments.

The Visible and Infra-red Survey Telescope for Astronomy, or VISTA, is a UK funded four meter class wide-field, infrared and optical survey telescope to be situated in Chile. The telescope, which can be regarded as a two-channel camera, provides a one-degree, infrared field and a two-degree, optical field. The project goal of low running cost requires minimal intervention by support staff. The optical configuration of the telescope must meet the scientific requirements while delivering a stable instrument profile for the 12-year duration of the survey. Accurate calibration of throughput, image quality and field distortion will be essential to the photometric and astrometric quality of that survey. The process for selection of optical design options is described and discussed with particular reference to meeting the functional and performance requirements, determined from the scientific specification, and to achieving a predictable and reliable image quality. The practical limitations imposed by budgetary and operational requirements are assessed for their role as contributory design drivers.

After the initial coating of the 8.3-m primary mirror of the Subaru Telescope in November 1998, we have conducted the first re-aluminization in August 1999. The primary mirror washing fixture worked efficiently for stripping the old coating and for washing the surface. Dry process is still to be improved. Suite of secondary and tertiary mirrors are being tested at the telescope, two of which were coated in- house, one in silver for the infrared observations and the other in aluminum for the optical observations, respectively. Evaluation of the coating film is conducted in two methods. Using a portable microScan, the reflectivity and the BRDF numbers of the primary mirror is monitored. Reflectivity over a wide range of the wavelength is measured in the witness mirrors. The preliminary data shows reasonably good number for the telescope optics. The in-situ cleaning of the primary mirror with solid and gaseous CO₂ sprinkle arms is operating once every month. Next step for the coating chamber commissioning is to improve the heating capacity for silver coating of the infrared secondary and tertiary mirrors, and the experiment for silver coating is going.

We report on the successful development of a new ultra lightweight optics technology. Intended applications include telescopes in space, on the Moon, and adaptive optics. The technology employs a novel process of optical replication using standard industrial grade graphite fiber composite materials. Composite replication makes possible telescope mirrors that combine extremely low areal density, large aperture, high surface smoothness, and high optical quality. Fabrication times and costs have been demonstrated to be far below that of competing lightweight optics technologies. The very low areal density achieved, ranging from 1 to 5 kg/m², makes possible multi-meter telescopes in space and on the ground. We present data on moisture absorption and outgassing, thermal expansion, thermal hysteresis, and improvement in optical figures. Applications to date include submillimeter telescopes and large optical arrays.

We present the conceptual design of a thermal control system for the KPNO Mayall 4-m primary mirror. An electrical current is passed through the reflective coating to match the temperature of the front surface of the mirror to ambient. Cooling of the back with thermo-electric coolers provides a cold reservoir for when the front surface temperature must be decreased. Temperature response of the front is estimated from a 1D thermal study: a 1C increase can be produced in approximately 1 hour from 25 wm^-2 of heating and a 1C drop in approximately 1 hour with a 10C temperature gradient front-back. Finite-element mechanical analysis is used to model optical affects and stress induced by a 10C temperature gradient through the mirror. Thermal print-through and mechanical stress are found to be negligible. The bending of the mirror by the front-to-back gradient is small and can be removed as a focus shift. In another study, modeling of heat transfer through free air convection simulates the response of a mirror without thermal control. For this case, the time to change the top surface temperature is long because of the mass of the mirror and the relatively poor transfer of energy by convection. For a hot mirror in a 5C cold environment to decrease the temperature by 1 C we estimate approximately 5 hours for the top temperature and approximately 24 hours for the core temperature. Image degradation from radial temperature distributions arising from convective heating or cooling of the mirror by surrounding air can become noticeable (0.05 - 0.1 arcsec FWHM, added in quadrature).

In this system, the segmented-mirror consists of three submirrors. They are hexagonal with 250 mm diagonal. The shape of each submirror is spherical with 3000 mm radius and the departure of each radius from the mean radius is less than 0.025 mm. A special S-H test device is used for co- focus measurement. It includes nine sub-apertures in three groups and each group is for a submirror. There are six displacement actuators in this system. The based unites of it are flexure hinges. The actuator is driven by a stepping motor with ten subdivisions. Six capacity displacement sensors are used in this experiment system. It is made in Tianjin University. A computer is used for data collecting, calculating and controlling. A special method for co-focus is developed in our work. By using this method the error of co-focus, i.e. the tilt error of submirror, is less than rms 0.035 arcseconds. The methods of calibration and maintaining for co-phase are also introduced in this paper. After once calibration, the diffraction limit image can be observed in about 220 mm aperture at (lambda) 650 nm, and it can be maintained about 20 minutes.

The Optical Very Large Array (OVLA) project consists in a kilometric-size optical interference of 27 mobile 1.5 m telescopes designed to provide high-resolution IR and visible snap-shot images. An OVLA prototype telescope has been developed at the Observatoire de Haute-Provence. It features a 1.5 m meniscus-shaped f/1.7 primary mirror weighting 200 kg including its active cell with 32 actuators. The mirror blank made of 24 mm-thick ordinary window glass is very cheap but extremely sensitive to temperature variations because of its large CTE (3 times larger than Pyrex). Indeed, the mirror shows a Z11 equals 3150 nm rms wavefront error due to a 0.5 degree(s)C thermal gradient generated between its front and back side by an unbalanced heat dissipation towards the night sky and the ground. This spherical aberration, too large to be corrected by the actuators, is compensated by an uniform electrical current generated through the aluminum coating by 42 peripheral electrodes. We also describe the electrodes control hardware and present some results obtained during the first light of the telescope. Lastly, we propose a possible upgraded surface heating system to adjust thermally other optical aberrations.

As a part of the Acquisition and Guidance Unit for the Gemini project a light-weight, 50 cm flat mirror has been designed at the Fraunhofer Institute for Applied Optics and Precision Mechanics in Jena as a subcontractor of the Carl Zeiss Jena company. A light-weight design of the mirror and its mount was essential since the total mass of the whole assembly including the positioning system was limited to 50 kg while interferometric quality of the mirror surface was required for arbitrary orientation. The overall surface error was below 54 nm r.m.s. while 27 nm was achieved in the central part. The mirror was fabricated from low-expansion glass ceramics to avoid thermally induced deformations. By milling pockets into its rear surface the mass of the mirror was reduced by 70%. The mirror is mounted cinematically via six solid-state hinges to three steel levers. The levers are connected to the mount frame at their centers via ball-and- sphere joints. This arrangement determines the position of the mirror uniquely while it allows for the thermal expansion of the mount frame. The position of the mirror as well as its tilt around an axis perpendicular to the optical one may be controlled a precision of 20 micrometers and 3 arcsec, respectively. The tilt axis is driven directly by two high- torque motors. To avoid an excessive power consumption of the motors the torque of the mirror head to be compensated for by a counterweight mechanism. The mirror may be deployed into the optical path using spindle driven linear rails.

We have obtained the curvature signal from defocused images before and after the pupil image for a simplified segmented mirror model. We used white light interferometry in order to calibrate the relative piston difference between the segments. The first results of applying the Curvature Sensing method to measure this relative piston difference are presented.

The paper describes the mechanism made for the 140-mm secondary mirror of the VLTI Auxiliary Telescopes. This mechanism consists of an hexapod parallel manipulator, capable of orienting its mobile base along all six degrees of freedom by means of six independent linear actuators.

Silver has the highest reflectance of all of the materials, but it tarnishes in the presence of sulfides, chlorides, and oxides in the atmosphere. Also, the silver reflectance is very low at wavelengths below 400 nm making aluminum more desirable mirror coating for the UV region. We have found a way to prevent silver tarnishing by sandwiching the silver layer between two thin layers of NiCrN_x, and to extend the metal's high reflectance down to 200 nm by depositing the (thin) Ag layer on top of Al. Thus, the uv is transmitted through the thin Ag layer below 400 nm wavelength, and is reflected from the Al layer underneath. This UV-shifted durable coating provides a valuable alternative to the aluminum coating for telescope mirror coatings where high throughput and durability are important considerations. The throughput for a telescope with, say, six reflections from silver coatings is (0.97)⁶ equals 83% compared to (0.92)⁶ equals 60% for aluminum coatings, or 28% less. The use of silver coatings allows more photons to be collected by primary mirror. Aluminum also has a reflectance dip at 850 nm caused by inter-band transitions which is eliminated by placing the thin Ag layer on top. This paper describes a non-tarnishing silver coating having high reflectance down into the UV region. The average specular reflectance is 70% - 97% in the near-UV, 95% - 99% in the visible region, and >= 99% in the infrared region covering the total wavelength range 200 nm to 10,000 nm.

A 1.3 meter aperture Cassegrain telescope with a very wide flat field has been completed and is now producing images. The field of view is a stunning 1.7 degrees while the focal ratio is a very fast F/4. The telescope is located at the United States Naval Observatory Flagstaff Station in Arizona, USA.

We explore solutions for the optical design of the OWL 100-m telescope, and discuss their properties, advantages and drawbacks in relation to top level requirements. Combining cost, design, fabrication and functionality issues, and taking into account the scale of the telescope, we conclude that the requirements are best met with a design based on spherical primary and secondary mirrors. The combined active and adaptive correction capability envisioned for the telescope allows substantial relaxation of otherwise critical subsystems specifications. We elaborate on the telescope correction capabilities, including alignment and focusing, and derive the structure of the optical error budget.