Institute of Astronomy

36-Inch Telescope

The telescope was built in 1951-55 by the now-defunct firm of Sir Howard Grubb, Parsons & Co. at Newcastle-upon Tyne. It replaced a much older telescope of the same aperture, which was brought to Cambridge from South Kensington when the Solar Physics Observatory moved here in 1913. That telescope was returned to its owners (The Science Museum) before the new one was installed; the Director of the Observatories at the time (Professor R.O. Redman), who in his youth had made substantial use of the old telescope, always averred that it should never have left the Museum!

The 36-inch, which is thought to be the largest telescope in the country, has three possible focal stations. There is a prime focus with a focal ratio of f/4.5; the primary mirror is a paraboloid, so no corrector is needed to obtain good images on the optical axis. In practice the prime focus has been little used: the telescope is large enough to make access to the focus difficult from the side of the tube. The other possible foci are coude, with a choice of two focal ratios, f/18 and f/30. The coude arrangement is unusual inasmuch as the light beam is directed UP the polar axis rather than downwards: that permits the shorter focal ratio to be exceptionally short for a coude, and results in a focus at a level near to that of the telescope, which is somewhat convenient for a lone observer who needs to operate both the telescope and whatever auxiliary equipment is placed at the focus. On the other hand, the arrangement lacks part of the advantage of a conventional coude focus, which is often in a basement that enjoys good passive thermal stability (and, from the point of view of the observer personally, protection from wind and extremes of cold!). Until recently the f/18 focus has been the favoured option, but new equipment that for the first time utilizes the f/30 arrangement has now been brought into use. The f/30 focus is just within the dome, high up to the north of the telescope, and its use involves a further reflection. In the present application, that reflection takes place close to the focus, and the beam is turned vertically downwards by successive internal reflections within two right-angle quartz prisms cemented together. The initial image is re-imaged at a focal ratio of f/14.5 at the position required for the auxiliary equipment. A simple plano-convex quartz field lens is cemented to the exit face of the quartz-prism assembly to image the telescope aperture upon the re-imaging lens.

In the early years of its operation, the telescope was used to send starlight into a spectrometer where the light intensities in several wavelength regions. which were accurately defined by masks in the focal plane of the spectrum, could be inter-compared. The intention (only partly realized, owing to the previously unrecognized individuality of the various stars) was to obtain astrophysically significant information about the chemical abundances and atmospheric characters of the stars surveyed. Three successive spectrometers, of progressively increasing size, resolution, and sophistication, were used in that effort.

The third spectrometer was further developed, some 30 years ago, to measure the doppler shift in the stellar spectra observed with it. It did that by means of a much more elaborate mask in the focal plane: instead of having just a few windows to isolate discrete bands of wavelength that were separately measured, it had a mask containing hundreds of narrow windows placed so as to match absorption lines in stellar spectra, the light from all of them being measured together by a single photomultiplier. The position of the spectrum could be sensed, and its doppler shift thereby accurately measured, by scanning the mask in the wavelength coordinate and looking for the more or less dramatic decrease in light transmission that occurs when every window is occupied by its corresponding absorption line. The plot of transmitted light against displacement of the mask is the cross-correlation function of the mask with the star spectrum, and has a pronounced minimum at the position of register.

That instrument, the orginal 'radial-velocity spectrometer', was the first application of cross-correlation to radial-velocity (or, indeed, any other astronomical) measurement. The method has now been adopted almost to the exclusion of the previous procedure involving the measurement of the positions of individual absorption lines, and has revolutionized the radial-velocity field, allowing observations to be made with enormously greater precision and sensitivity than was possible before. A few years before the instrument was brought into operation, a compilation of all known stellar radial velocities included only about 70 stars of 7.0 magnitude or fainter whose radial velocities were supposed to be known to an accuracy of 1 km/s; more and fainter stars than that were sometimes observed to at least that accuracy on individual nights in Cambridge - a site that has not generally enjoyed much of a reputation for its excellence for observation. The original instrument remained in operation for 25 years, during which it provided most of the data for about 200 published scientific papers, and when it was de-commissioned it went straight to the Science Museum as an historic instrument. There were delays in commissioning its successor, which is however operating now and provides sensitivity, precision and convenience well beyond those of the pioneering instrument.

Additional information about the telescope and radial-velocity spectrometer is obtainable at any time from Dr. R.F. Griffin (rfg@ast.cam.ac.uk), who will also be glad to arrange informal demonstrations of the equipment both by day and at night.

Page last updated: 18 February 2011 at 12:54