An all-spherical catadioptric Gregorian telescope design suitable for meter class telescopes is described. A group of small lenses is placed at the intermediate focus of the Gregorian system to correct the aberrations of the f/1.5 spherical primary mirror and spherical secondary mirror. The corrective lenses are of simple form and conventional substrate. The system's performance is ideal for broadband CCD imaging and is diffraction limited in the visible portion of the spectrum.

Much effort has been expended in eliminating the need for aspherics in reflective telescope designs. Typically, the imperfect imaging properties of spherical mirrors are best corrected with lenses at or close to the pupil of the system, or in other words, full aperture correctors. The massive lenses become optically, mechanically and thermally impractical when the aperture of the system exceeds 0.4 m. Sub-aperture correctors, when applied to conventional Cassegrain and Newtonian systems with spherical mirrors, have not met with much success since the residual color aberrations typical of such systems are undesirable.

The advances in computer aided optical fabrication and active telescope technology have tamed the difficulties associated with the production and use of aspherics in very large observatory telescopes. However, exploring the elimination of aspherics in smaller, meter class telescopes is still worthwhile in order to bring costs down to a level affordable by smaller institutions and advanced amateur astronomers.

The catadioptric Gregorian reflector design described here is believed to be a novel approach to an all-spherical optical system that is both simple and relatively inexpensive to produce optically. The design is diffraction limited over most of the visual spectrum and the image size over a very broad spectral band lies within a 1 arc second circle. The overall dimensions of the system's optical tube assembly are similar to those of a conventional Cassegrain telescope due to the high-speed, f/1.5 primary mirror. The Gregorian optical configuration also offers superior baffling of stray light to that of a typical Cassegrain layout.

The all-spherical catadioptric Gregorian design was inspired by the desire to build a large telescope, and the availability of a 1 meter f/1.5 mirror blank. The use of an aspheric primary was ruled out due to the difficulty of figuring such a large and fast mirror with conventional fabrication techniques, not to mention the significant problems associated with collimating the finished telescope.

The challenge of eliminating the primary's considerable spherical aberration in a simple system seems daunting, but this feat is accomplished regularly in optical shops when null test techniques are applied to aspheric mirrors. The spherical aberration correction in the catadioptric Gregorian telescope design, illustrated below, was inspired by the Offner null test technique. While the Offner null corrects the spherical aberration of a mirror at the center of curvature, in this adaptation it corrects the aberrations of a spherical telescope mirror at its infinity focus and over an extended field.

Of a two-lens form, the Offner null's small positive relay lens corrects most of the primary spherical aberration, while the field lens close to the center of curvature contributes to the correction of higher order spherical aberration residuals.

Since null tests are performed using monochromatic light, chromatic aberration is not an issue. Adapting the null lens concept to a white-light imaging telescope necessitates replacing the positive relay lens with a mirror, since color dispersion by this element would be overwhelming. The resultant optical system now resembles a Gregorian configuration with the field lens at the intermediate focus. The corrector lens diameters are constrained to fit within the centrally obscured region of the imaging rays.

The initial design of the system, once the basic idea was conceived, was surprisingly easy. A Gregorian mirror system was laid out and a few weak positive lenses were "tossed in" around the intermediate focus. The Zemax default merit function was used, and after several optimization runs the design quickly converged toward a system that provided reasonable images, although the lens forms were far from ideal. Further design effort revealed that excellent performance could be achieved with as little as four small corrective lenses of conventional glass types.

The f/8.5 design presented here is diffraction limited in the central visual part of the spectrum using only four corrective lenses of conventional glass types. The image size over the broad spectral range of 360 nm to 1 micron is no larger than 1 arcsecond (corresponding to a 2x2 array of 24 micron pixels) across much of the central area of the detector's field, with degradation occurring in off axis images only at the extreme ends of the spectrum. The spot diagrams illustrate the performance of the system over a 1k square CCD array of 24 micron (0.55 arc second) pixels. An all-silica corrector lens design has also been formulated to access a broader range of the spectrum, from 340 nm to 1 micron, more efficiently. To preserve the image quality an extra element was added before the intermediate focus.

A thorough tolerance analysis has yet to be completed. However, a sensitivity analysis, and even a simple visual examination of the layout alone, suggests that construction of this system will pose some challenges. Light rays converge upon the corrector group and secondary mirror at steep angles, suggesting the alignment of the secondary with respect to the corrector lenses will be critical. The lenses themselves do a significant amount of "work" to compensate for the primary's aberrations, and therefore will be sensitive to manufacturing and mounting errors, hence requiring a carefully designed and fabricated cell. Aligning the spherical primary with the secondary-corrector lens assembly is straight forward, as it has no discrete optical axis, however the primary's radius of curvature must be very tightly controlled in fabrication since a small variation will result in a significant change in its spherical aberration contribution.

The telescope will require two sets of spider supports. The spider supporting the corrector lens cell can be mounted in a tube that also holds the secondary mirror. This tube would also serve as one of the baffles for the optical system, as illustrated below. The other spider support will suspend the secondary/corrector assembly in the main tube. In order to minimize the number of diffraction spikes, the spider vanes will have to carefully aligned.

While initial exploratory optical design work has been done on the all-spherical catadioptric Gregorian telescope, a thorough tolerance and ghost image analysis has yet to be performed. Nor has a prototype been fabricated. Builders should be cautious, as there may be unforeseen problems with this design in practice. The design prescription is offered here to encourage others to explore this design form further.