The manufacture of large single crystals for optical applications presents many challenges to crystal growers and processors. Although the general trend for optical device sizes is to produce smaller units, there is still demand for producing crystal materials for large aperture optics such as windows and lenses. These applications continue to become more demanding in terms of size, homogeneity and surface finish and there are many areas of manufacture which require a significant amount of skill rather than scientific input.

Schott had delivered blanks for large lenses and prisms since many decades. Glass and glass ceramics objects with dimensions above 300 mm diameter or edge lengths will remain challenges for a glass manufacturer. This holds especially when the quality specifications exceed the standard level significantly. The developments in glass manufacturing allow casting of ZERODUR^R glass ceramic blanks up to about 1.5 m with homogeneities like that of high grade optical glass. This was utilized to test a convex mirror using a large Zerodur element in transmission providing the concave interferometer reference surface and the imaging aspherical lens simultaneously. Optical glass blocks of more than half a ton have been produced with outstanding internal quality. Although the manufacturing process is well controlled there are restrictions on the availability of such objects (glass types, long process times e.g.). The glass production process is presented pointing out its implications as a guideline for designers in order to avoid unneccessary time losses.

A fiber-fed spectrograph (HERCULES) has recently been installed at the Mt. John Observatory. This spectrograph achieves resolving powers of 35,000 and 70,000 using 100 micrometers and 50 micrometers fibers. The spectrograph is designed to capture the visual spectrum from 380 nm to 880 nm in a single exposure on a 50 mm square CCD. As the spectrograph is enclosed in a vacuum tank and is thermally isolated it is expected that high precision radial velocity measurements will be obtainable. Overall instrument efficiency is predicted to peak at 23% in 1' seeing. Preliminary design ideas for the 11 m Southern African Large Telescope (SALT) high resolution spectrograph are also discussed. While it is hoped that a variant of the HERCULES design can be adopted for this instrument, as with all spectrographs designed for large telescopes, the design presents significant challenges.

Two different approaches are discussed for computing the expected ELMER image quality. The final Error Budget is computed in two different ways: a hierarchical model and a sequential model. The hierarchical model is the classical process for error budget building, in which the requirement degradation due to each error source is analyzed independently, combining all errors at the end. The sequential model follows the real design and fabrication process, simulating the steps sequentially. The results obtained in each step feed the following one, degrading the requirements from the Encircled Energy Radius (EER) design value to the final one. Finally, both results are compared.

EMIR is a multiobject intermediate resolution near infrared (1.0 - 2.5 microns) spectrograph with image capabilities to be mounted on the Gran Telescopio Canarias (Observatorio del Roque de los Muchachos, La Palma, Spain). EMIR is under design by a consortium of Spanish, French and British institutions, led by the Instituto de Astrofisica de Canarias. This work has been partially funded by the GTC Project Office. The instrument will deliver images and spectra in a large FOV (6 X 6 arcmin), and because of the telescope image scale (1 arcmin equals 52 mm) and the spectral resolution required, around 4000, one of the major challenges of the instrument is the optics and optomechanics. Different approaches have been studied since the initial proposal, trying to control the risks of the instrument, while fitting the initial scientific requirements. Issues on optical concepts, material availability, temperature as well as optomechanical mounting of the instrument will be presented.

We report on the status of an on-going project for the realization of two prime focus imaging cameras for the Large Binocular Telescope. Each channel is optimized for a specific wavelength range, namely the blue and the red portions of the visible spectrum. One of the most challenging parts of the instrument, concerning optical design and mounting issues, is represented by the optical correctors required to balance the aberrations induced by the parabolic primary mirrors, over a 0.5 degree full field of view.

The new wide-field infrared camera (WFCAM) for the 3.8 m United Kingdom Infrared Telescope (UKIRT) and the Visible and Infrared Survey Telescope for Astronomy (VISTA) are two future facilities to provide sky survey data to complement the latest 8 m-class telescopes. Both use large lenses (with a diameter greater than 500 mm) made of fused silica. The cryostat window used for the IR channel in VISTA is 950 mm in diameter. The specification and cryogenic mounting of these components, some of which with aspherical surfaces, are an opto-mechanical challenge and are described in this paper. We also discuss various aspects of the manufacturing and testing.

The design of large lenses, say 0.2 to 1.0 meters in diameter, is potentially an area for development, but progress is sporadic. Designs for large lenses, complicated as they are by the optical properties of glass, are more difficult than those for reflecting systems. This paper makes an initial survey in which various problems of large lenses are outlined. An extreme, historical problem in the design of large lenses is re-examined as a basis for moving forwards in future applications. Examples of modern large lenses are briefly discussed.

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, which is almost fully funded and whose construction has begun at the time of writing, 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 (SAC) mounted on a tracker beam at the prime focus delivers a high quality image to the focal plane and enables a celestial object to be followed for twelve degrees across the sky. The SAC is the subject of this paper: the results from the already published design study will be summarized, and recent work on the system that will be built will be presented.

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, which is almost fully funded and whose construction has begun at the time of writing, 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 (SAC) mounted on a tracker beam at the prime focus delivers a high quality image to the focal plane and enables a celestial object to be followed for twelve degrees across the sky. The telescope is to be optimized for ultraviolet wavelengths so it is especially important that it is equipped with an atomospheric dispersion corrector (ADC). This paper will present a concept for an ADC for SALT using two large prisms whose variable separation can be adjusted for the different zenith distances of the target during observation. This kind of ADC is in use on the ESO VLT and is planned for use on SOAR.

Many lighthouse services tod ay are removing or decommissioning traditional optics and installing new, smaller, self-contained devices. There are good economic reasons for doing this but sometimes there is a need to retain large traditional optics. In this case the choice of light source is important, fitting a modern 'off the shelf' lamp in a large optic can produce poor results. Over the last three decades light intensity measurements of several lighthouses have been carried out in order to determine their performance. During the course of this work several experimental light sources have been measured and their results compared with existing light sources. This process started in 1971 with the first measurement of a lighthouse fitted with a paraffin vapor burner. Several experimental light sources were temporarily installed and compared with existing light source. Since then experiments have continued and resulted in the replacement of 3.5 kW lighthouse lamps in rotating optics with readily available 1 kW metal halide lamps. Within recent years, in a drive to reduce energy requirements for potential use of solar power, some small low power lamps have been temporarily installed in large optics to see how they performed. In many cases the small size of the light source has caused problems of poor performance including short flash duration and low intensity. Various techniques such as envelope etching, reeded diffusion and lamps clusters have been used to enhance the performance of these low power light sources in order to optimize their use within traditional optics. The results of various light measurements are shown in this paper, together with details of problems encountered during the experiments.

Widespread adoption of precision and ultraprecision articles from the polymeric materials creates a need for the understanding of a mechanism of the new high quality surfaces generation by the controlled fracture processes in the single-point diamond machining. The efficacious way for this understanding is a creation of the model of the surface layer forming process as result of the formation of its accidental and methodical defects by the precision microcutting.

This paper describes progress on the development of a new process for producing precision surfaces for the optics industry, and potentially for other sectors including silicon wafer fabrication and lapping and polishing of precision mechanical surfaces. The paper marks an important milestone in the development program, with the completion of the construction of the first fully-productionized machine and the first results from the commissioning process.

Over the last decade, we have witnessed that the fabrication of 200 - 2000 mm scale have received relatively little attention from the fabrication technology development, compared to those of smaller than 200 mm and of larger than 2000 mm in diameter. As a result, the optical surfaces of these scales are still predominantly completed by small optics shops where opticians apply the traditional technique for polishing. Lack of tools in aiding opticians for planning, executing and analyzing their polishing work is a root cause for long and, sometimes, unpredictable delivery and high manufacturing cost for such optical surfaces. We present the on-going development of a software simulation environment called Surface Analysis and Fabrication Environment (SAFE). It is primarily intended to increase the throughput of polishing and testing cycles by allowing opticians to simulate the resulting surface form and roughness with input polishing variables. A brief review of current polishing techniques and their target optics clarifies the need for such simulation tool. This is followed by the development targets and a preliminary simulation plan using the developmental version of SAFE. Among many polishing variables, two removal assumptions and three different types of removal functions we used for the polishing simulation presented. The simulations show that the Gaussian removal function with the proportional removal assumption resulted in the fastest, though marginal, convergence to a super-polished surface of 0.56 micron Peat- to-Valley in form accuracy and of 0.02 nanometer in surface roughness R_a. Other meaningful results and their implications are also presented.

Aspherical surfaces on lenses are difficult to produce and test. There are many innovative approaches used today to address the complexities of this process. The size, quality, and quantity of the lens to be produced dictate the best approach. The UCO/Lick Observatory Optical Lab has fabricated aspheric lenses to diameters over 12 inches using fabrication and testing techniques developed specifically for high quality, one-of-a-kind lenses, with axi-symmetric profiles and significant departures from sphere. This paper describes the manufacturing procedures used at UCO/Lick to fabricate aspheric lenses typical for today's astronomical applications.

Lenses made from the optical crystal, calcium fluoride, are required in spectrograph camera designs for many of today's astronomical telescopes. Often these cameras are of large aperture with lens diameters in the 12 inch to 15 inch range. This material is highly susceptible to mechanical and thermal shock, resulting in fracturing. Large calcium fluoride lenses, with their varying thickness profiles, require techniques and precautions in processing from raw material to a finished lens that aren't required with windows or small lenses. This paper describes techniques learned and used at the Lick Observatory Optical Lab that deal with working this delicate crystal and producing large finished lenses.

The challenges of fabricating large prisms are considered. Production variable properties of glass that affect the optical performance of an optical system are briefly considered. Techniques used to manufacture a large Penta prism are described.

The cementing in 1974 of the two glass components of what was the world's largest cemented doublet for the UK Schmidt telescope will be described. The subsequent final polishing and testing will be reviewed and some of the lessons learned during the cementing procedure reported.

To present a full account of the developments in the manufacture of large lenses one needs to address wider issues rather than just provide a catalogue of technological progress. The advances in glass manufacture and improvement in optical techniques have to be considered in relation to the cultural, social and economic factors that have determined where, how and why large lens manufacture developed in specific countries. The challenge facing historians trying to tackle this technological theme, is that it is often poorly documented and little is preserved in the historical records. Until relatively recent times, opticians have concealed their methods, trade secrecy being an important economic strategy. To provide an example, it should be noted that although William Herschel produced the best optics and telescopes of the day, he published practically nothing about his methods. What has been gleaned of his techniques has only been uncovered by careful study of surviving manuscript sources and measurement of his surviving optics. Such was William's personal knowledge, that his son John had to take instruction from his father to refurbish William Herschel's 20-foot telescope. This training gave John tacit knowledge of William's methods and allowed him to successfully undertake his cape observations in the Southern Hemisphere. In spite of the shortcomings of the historical record, historians can give a measured account of the developments of lens optics by studying surviving telescopes and their optics.

In this progress report, the authors discuss the use of a laser beam, propagating freely in air, as a reference line for profilometry with a large travel length. The reference beam is detected by a position-sensitive detector and any variation is compensated physically by a two-axis flexure system in closed loop operation. Air turbulence is the main obstacle for the stability of the reference system. It is shown that the relative positions of the position-sensitive detector and the height measurement system is critical for correcting the rotational error of the carriage.

Traditionally, the key component design parameters such as radius, lens thickness, size and shape of most types of optical components are measured using optical techniques. There are several reasons for this, but in particular: the form of the entire surface is generally revealed in one testing set up, the optical functionality of the component is almost always the performance defining factor, and the heritage of many of the optical methods, i.e. the huge investment and expertise that has in the past been brought to bear on perfecting the testing methods. There are however, alternative non-optical instruments, such as CMMs, for measuring the form of optical components that are becoming increasingly attractive for conformal optical components, off-axis optics, and aspherical lenses and components, for example corrector plates, grisms, and so on. The main reasons for the increased acceptance of such techniques are that: the asphericity of some of the surfaces is often too great to be handled satisfactorily by interferometric methods at optical wavelengths, or even at infra-red wavelengths; the probing force of modern, special- purpose probes is remarkably small; the cost of producing computer-generated holograms required for optical testing can be very high and often the numbers of components to be tested do not justify the expense; the speed of production is such that the component cannot be repeatedly removed and replaced in the manufacturing machine and/or the manufacturing process is not so conductive to optical testing because of the presence of cutting fluids etc. and, finally, the level of accuracy required cannot be achieved using optical techniques for unorthodox shapes.

Due to the demanding specifications required by the modern optics industry a large number of instruments and methods have been developed to establish an appropriate metrology infrastructure. When measuring surface texture (profile, waviness and roughness), the form of the test object is removed, either by some type of mechanical skid device or by digital filtering techniques. In the field of large optics, deviations from ideal form can be many orders of magnitude larger than the surface texture specifications - this makes the metrology difficult. The most common forms of instrument used to measure optical surfaces are based upon non-contact optical probing techniques for which there are no international specification standards. These problems, along with the multitude of pick-up systems available, mean that it is difficult to obtain traceability to the definition of the meter. This paper discusses the issues that plague the surface metrology community and presents the results of a number of comparisons.

White light scattering interference, first observed by Newton, generates wide aperture fringe patterns permitting identification of the position of zero order interference. High autonomous fringe contrast level was noted and is explained and optimum scatterer density quantified. Applications to large lenses and prisms are considered.

The Megajoule laser (LMJ) and its first prototype, the Laser Integration Line (LIL), is equipped with a specific final optics assembly involving two diffraction gratings instead of a classical focusing lens. Both gratings have a dimension of 420 X 470 mm² and are working at an incidence of 25 degree(s).. Gratings are plano transmission holographic gratings directly engraved into fused silica substrates. The 1(omega) grating is working at the wavelength of 1.053 micrometers , its grooves are straight and equispaced. The 3(omega) grating, is a focusing grating working at the wavelength of 0.351 micrometers . Its grooves are curved and non equispaced. The gratings were designed and manufactured to present efficiencies superior to 90% on the whole clear aperture and an improved damage threshold. After Jobin Yvon's selection by CEA in 1999, specific equipment and facilities were put in place to manufacture these large gratings. The aim of this contribution is to present the early results of the development of this 1(omega) and 3(omega) gratings. After a short introduction to the 1(omega) and 3(omega) gratings specifications, manufacturing process, efficiencies result and AFM profiles of the first manufactured gratings will be detailed.

The design of high power lasers such as the MEGAJOULE laser (LMJ) and its first prototype, the Laser Integration Line (LIL) requires optical components with very strict and diverse specifications over large apertures. Though technologies used for the fabrication of these components may be usually compatible of such specifications, fabrication processes are often restricted by our ability to measure the effective performances. In order to determine the effective quality of its components and to help optimizing their production, CEA equipped with a wide range of metrology devices, many of them were developed for the specific needs of LIL and LMJ programs. In the same time CEA also supported the development of specific metrology means at its optics vendors to help in th fabrication process design. After a short description of the Megajoule laser, we will focus on the different metrology devices used in the characterization of its optical components non-exhaustively ranging from interferometry and photometry measurements to focal spot analysis.