Making the mirror surfaces

The largest single mirror that has been produced by polishing is the 8.4-m mirror installed at the University of Arizona in the United States. There is no plan to make a larger mirror, as it is commonly understood around the world that future large telescopes can only be produced through use of segmented mirrors. All telescopes planed to have a primary mirror larger than 10m adopt ideas of segmented mirrors, although there is variety in sizes and shapes of planned segmented mirrors. The major problems encountered in use of segmented mirror systems are associated with control technologies for making non-axisymmetrical mirror surfaces and ensuring that mirror segments function as one large mirror when assembled. However, these technologies have been demonstrated to be feasible in the Keck 10-m telescope mirror (consisting of 36 segments) in the United States, showing that they can be used for future extremely large telescopes.
Japan thus far only has experience in producing up to 1-m-class mirrors. In terms of international competitiveness, Japan is a long way behind in mirror production technology. However, for extremely large telescopes, if technology for making many small, non-axisymmetrical mirror segments rather than building a single large mirror can be developed, then the gap will be narrowed and Japan will be able to compete with the rest of the world on equal terms.
The primary mirror of a 30-m-class extremely large telescope will consist of approximately 1000 axially asymmetric, aspherical mirror segments of approximately 1m in size. These elements will be assembled to form a large mirror. However, if traditional polishing technology is employed, a single polishing machine will require one year to prepare all the element mirrors, at a cost of 50 million JPN. Furthermore, using normal polishing methods it is only possible to make axially symmetric mirrors; a number of extra production steps need to be introduced in order to make a segmented mirror. Therefore, the present plan has tended toward using grinding rather than polishing as the leading candidate mirror-making technology.