The James Clerk Maxwell Telescope

The James Clerk Maxwell Telescope (JCMT)^2.4 is the world's largest sub-millimetre radio telescope. As it is presently the telescope that is most likely to produce data that the programs described in this thesis will attempt to model it will be used here to describe the way in which a typical sub-millimeter radio telescope collects radiation and produces the end data product. Located 4092m above the Pacific Ocean at the summit of one of the world's highest volcanoes (Mauna Kea, Hawaii) it has a 15m main receiving dish which is housed within a 22m high and 28m diameter cylindrical shaped housing. The housing is designed to protect the telescope from the sometimes very harsh conditions on top of the mountain. When the telescope is to be used the housing is opened but the telescope still remains protected behind a large gore-tex shield which protects it from dust and wind. The membrane is 80% transparent to radio waves but reflects visible and infrared light thus enabling the telescope to be pointed at the sun for observations if desired. The dish is a Cassegrain design which reflects the received radiation via a 0.75m secondary into a receiver cabin mounted directly behind the dish. In order to accurately receive radiation at sub-millimeter wavelengths the dish is manufactured to an accuracy of about 30$\mu$m and consists of many adjustable panels which in theory enable the optimum beam shape to be formed. The received radiation in the receiver cabin is then directed by a movable mirror to one of several instruments in the receiver cabin or out to a platform on the end of the elevation bearing - this is used for instruments that are either too large or too heavy to go in the receiver cabin or instruments which are not designed to be rotated (as the receiver cabin moves with the telescope). There are two primary types of receiver used on the JCMT, heterodyne receivers which are used to observe spectral lines and photometric receivers for continuum observations. As this thesis is concerned only with spectral line observations only the heterodyne receivers will be described here.

All the present receivers on the telescope use a Superconductor-Insulator-Superconductor (SIS) junction to receive the radiation. These devices are cooled to 4K in order for the superconducting properties to come into effect^2.5 and also to reduce the thermal noise to the lowest possible level. As the frequency of the radiation being received is very high (200-900 GHz) it is not possible to build amplifiers to deal with these frequencies. Therefore the signal is mixed with a reference signal from a local oscillator (typically about 100 GHz) to produce an Intermediate Frequency (IF) which is at a low enough frequency (about 1-3 GHz) to be handled by the electronics. This mixing of the signals produces two so called passbands within whose frequency range the radiation is detected. These are referred to as the lower and upper side bands and are separated by $2\nu_{\rm IF}$ where $\nu_{IF}$ is the IF. Unfortunately in most receivers it is not possible to separate these two frequency ranges and the end result from the receiver is one band which consists of the sum of the two passbands. This can make line identification rather difficult as it may not be certain in which sideband a particular line lies.

In order to be able to measure the strength of a line the receiver must be calibrated as accurately as possible. This is done by placing in the beam of the receiver a 'hot' and a 'cold' temperature source. The hot source is generally the ambient room temperature and the cold source is usually the outer cooling layer of the receiver which is generally liquid nitrogen at 77K. The signal generated by these sources provides a scale against which the actual signal from the astronomical source can be calibrated.

Since the telescope is located within the Earth's atmosphere it is viewing the astronomical sources through a radiation source and it is necessary to remove this signal from the output. This is done by one of two methods, position switching or frequency switching. Both methods assume that the properties of the atmosphere do not change significantly with small position changes or small frequency shifts. Thus when using position switching the telescope is pointed at the source in the sky for a period of time and the signal collected, the telescope is then moved (or actually the secondary mirror moves slightly) to point at a different piece of sky which does not contain the source (but is still close by so the beam is effectively passing through the same piece of atmosphere) and the signal is collected here for the same length of time. The two signals are then subtracted from one another and thus the sky noise should be removed and all that is left will be the signal from the astronomical source. Frequency switching works in a similar manner except the frequency is shifted slightly off the desired frequency rather than the position being moved. Frequency shifting has the disadvantage that it tends to produce very poor baselines so position switching is the method of choice.

The final stage of acquiring the data is to send the output from the receiver to some electronics to convert the output into usable data. All the spectral line receivers use a rack of electronics called the Digital^2.6 Autocorrelation Spectrometer (DAS). The DAS splits the output from the receiver into up to 2048 different channels and measures the strength of each channel. It is highly configurable and has a wide range of available channel widths that can be chosen. The output from the DAS is sent to the JCMT computer system which then stores the intensity for each channel. This can then be displayed as a typical spectrum.