The seven telescopes in VERITAS will be identical and will have the geometrical layout shown in Figure 11. Six telescopes will be located at the corners of a hexagon of side 80m and one will be located at the center. The telescopes will each have a camera consisting of 499 pixels with a field of view of 3.5diameter. The design of the individual telescopes will be similar to that of the existing Whipple 10m reflector which has proven to be an excellent facility for VHE gamma-ray astrophysics. It has withstood mountain conditions for over thirty years, has good optical quality, a structure sufficiently rigid that flexure is negligible at all elevation angles, and it has good UV reflectivity.
The Davies-Cotton optical design ([Davies & Cotton 1957]) used in the 10m reflector (and many subsequent gamma-ray telescopes) has facet mirrors which are spherical and identical, facilitating fabrication at a reasonable cost and making alignment easy. It also has smaller off-axis aberrations than a parabolic reflector so that it has good image quality out to a few degrees from the optic axis. Its one limitation, that the surface is not isochronous, means that the 10m reflector introduces a time-spread of 6ns into the light pulse. We, as well as other research groups, have investigated alternative designs but conclude that the Davies-Cotton collector provides the optimal combination of optical quality and cost effectiveness. Increasing the f-number from f/0.7 to f/1.2 substantially improves the optical quality of the telescope (Figure 12). It is then possible to match the inherent fluctuations in the shower images with a camera that has a reasonable number of pixels ( 500) and appropriate field of view (3.5).
The desirable optical parameters of the telescope follow from the characteristics of a gamma-ray shower Cherenkov image. For a gamma-ray source at the center of the field of view (FoV), the average image centroid (distance) is at about 0.85 from the center of the focal surface, and moves to slightly larger values at larger energies. The image has a r.m.s. width and length of about 0.14 and 0.25, respectively. The reflector will have sufficient resolution to record structure on this scale. The intrinsic time width of the Cherenkov light is a few nanoseconds and the spread in light across the f/1.2 reflector is 3-4ns.
A key factor in the reflector quality and cost is the mirror facets. Although a number of mirror materials and fabrication technologies have been developed and investigated for this project, a conventional approach has been chosen as the most cost effective. The facets will be made of float glass, slumped and possibly polished, aluminized and anodized. They will be hexagonal in shape to permit close packing and will be 60cm across the flats.