Optical Telescope Assembly (OTA)

Optically, the Hubble is a Cassegrain reflector of Ritchey-Chrétien design, as are most large professional telescopes. This design, with two hyperbolic mirrors, is known for good imaging performance over a wide field of view, with the disadvantage that the mirrors have shapes that are hard to fabricate and test. The mirror and optical systems of the telescope determine the final performance, and they were designed to exacting specifications. Optical telescopes typically have mirrors polished to an accuracy of about a tenth of the wavelength of visible light, but the Space Telescope was to be used for observations into the ultraviolet (shorter wavelengths) and was specified to be diffraction limited to take full advantage of the space environment. Therefore its mirror needed to be polished to an accuracy of 10 nanometres, or about 1/65 of the wavelength of red light.

Perkin-Elmer intended to use custom-built and extremely sophisticated computer-controlled polishing machines to grind the mirror to the required shape.However, in case their cutting-edge technology ran into difficulties, NASA demanded that PE sub-contract to Kodak to construct a back-up mirror using traditional mirror-polishing techniques.(The team of Kodak and Itek also bid on the original mirror polishing work. Their bid called for the two companies to double-check each other's work, which would have almost certainly caught the polishing error that later caused such problems. The Kodak mirror is now on permanent display at the Smithsonian Institution. An Itek mirror built as part of the effort is now used in the 2.4 m telescope at the Magdalena Ridge Observatory.

Construction of the Perkin-Elmer mirror began in 1979, starting with a blank manufactured by Corning from their ultra-low expansion glass. To keep the mirror's weight to a minimum it consisted of inch-thick top and bottom plates sandwiching a honeycomb lattice. Perkin-Elmer simulated microgravity by supporting the mirror on both sides with 138 rods that exerted varying amounts of force. This ensured that the mirror's final shape would be correct and to specification when finally deployed. Mirror polishing continued until May 1981. NASA reports at the time questioned Perkin-Elmer's managerial structure, and the polishing began to slip behind schedule and over budget. To save money, NASA halted work on the back-up mirror and put the launch date of the telescope back to October 1984.The mirror was completed by the end of 1981; it was washed using 2,400 gallons of hot, deionized water and then received a reflective coating of aluminium 65 nm-thick and a protective coating of magnesium fluoride 25 nm-thick.
 
Doubts continued to be expressed about Perkin-Elmer's competence on a project of this importance as their budget and timescale for producing the rest of the OTA continued to inflate. In response to a schedule described as "unsettled and changing daily", NASA postponed the launch date of the telescope until April 1985. Perkin-Elmer's schedules continued to slip at a rate of about one month per quarter, and at times delays reached one day for each day of work. NASA was forced to postpone the launch date until first March and then September 1986. By this time the total project budget had risen to US$1.175 billion.