The large telescopes that have come into being in the last thirty years or so are some of the engineering marvels of the world, with a precision untouched in few other disciplines. Large telescopes are not new. Extremely long ones were used in the 17th and 18th centuries, usually to use the centre of what lenses were available and to avoid chromatic aberration in refracting telescopes. The 19th century saw the rise of precision engineering and large refractors were built in a number of places, some of which are still around today. The biggest is the Yerkes Observatory 40 inch refractor, which still works. This is as large as you can take a refractor – glass is not entirely solid and lenses tend to sag and lose their accuracy, however they were ground in the first place.
All big telescopes are reflectors, often of classical designs such as Cassegrain or Gregorian, but the favoured conformation is Richey- Chretien, where careful design can eliminate many aberrations. The 200-inch telescope on Mount Palomar, commissioned just after World War 2, is supported on an equatorial mount, but the engineering involved is such that all subsequent larger telescopes are fork (or alt-az) mounted. In a sense the 200 inch was the last of an era, and a magnificent machine it is, still in use. Mirrors are now aluminised, not silvered, and the accuracy is often down to fractions of the wavelength of light.
I am going to use as my example the two Keck Telescopes on their mountain in Hawaii, 4000 metres
or so in elevation.
They are not new, but they illustrate most modern features, including the ability to be used together to create some interferometry. The main mirrors on the telescopes are 10 metres in diameter,
and instead of one piece of glass they are made of 36 segments joined together at the edges to create the desired curvature. In the case of Keck they are hyperbolic mirrors (when joined up). The mirrors are much thinner than on Mount Palomar and so special measures are needed to ensure that they retain their shape. (In parenthesis it is possible to cast and figure large thin mirrors but only on lab does it, in Arizona, and a number of these 8 metre mirrors are going to be put together to form an enormous machine that will come on stream in the next decade.)
To keep the mirrors in the correct shape a system of actuators is employed beneath the mirror bed. A constant set of readings are taken and fed into a computer which passes corrections to the activators. It is possible to keep the mirror correctly shaped to nanometre tolerances, and the activators work several times a second. This system is known as Active Optics, and it has been retro-fitted to many telescopes. This system solves one problem, that of distortion of shape, but it does not correct for the twinkling of light that comes as the light from the star interacts with the atmosphere of the earth. Without correction the resolution theoretically possible with a huge telescope would not be realised – the image is twinkling. Adaptive Optics is used to address the twinkling problem. By the use of a small rapidly tilting mirror (tip-tilt) and also small distortable lenses the degree of “twinkle” can be compensated for and a clean image obtained. Needless to say such a system relies on feedback loops administered by some serious computer power, and Adaptive Optics operate several thousand times a second! The results are nothing short of remarkable – images as good or better than Hubble can be obtained from earth, and resolutions measured in sub arc seconds. Such a system needs a bright star for activation, and in the usual absence of higher magnitude stars laser beams are employed.
All modern telescopes use such systems, and the Kecks are notable only in being among the first in the field and being very good instruments.
There are many problems associated with these telescopes, cost being one. The Kecks cost 192 million $; the latest European venture has been capped at 1.1 billion euros. They have to be set on mountains, often very remote – there is no point in putting such a machine in the middle of London. They have to be run at low temperature, so huge air- conditioners are installed. It is less of a problem with the very large aperture telescopes but some of the sky survey scopes produce so much data that storage is a problem. 1015 bytes of data per annum is not unusual, which will take decades to analyse. (Hence the advent of amateur data analysis, which has been successful)
The giant scopes are a tribute to the engineering skills of their manufacturers, but also to the ingenuity and creativeness of their designers.