•■A':-, ™

Fig. 6.15. Spectrum of an absorption line in the presence of baseline ripple (lower frame). In the middle frame there are two observations whereby the feed has been axially defocused by + and - A/8. Adding these to form the upper trace shows that the ripple with 300 km/s "period" has been suppressed. The wide ripple in the top frame is due to another multipath interference. (Pillai, MPIfR)

In the design of modern radio telescopes the avoidance or minimisation of these multiple reflections has been included in the design effort. As Morris showed, the following measures can greatly contribute to the suppression of baseline ripples:

i) path A: avoid reflections on structures near the feed by covering these with absorbing material. Of course the reflection on the feed itself cannot be avoided. Minimise the backscatter from the central region of the reflector to the feed by a "splash plate", directing the radiation outwards. In a Cassegrain system this can be achieved effectively by shaping the central area of the subreflector into a cone. The size and angle of the cone are chosen such that the radiation is directed to the main reflector just outside the central hole which provides access to the secondary focus, where the feed is located.

ii) paths B and C; here it is important to avoid specular reflection on the support legs by shaping the legs with a sharply pointed wedge towards the main reflector. This will reflect the radiation away from the direction of the main reflector.

iii) in a Cassegrain system there will be some edge diffraction on the subreflector. This diffracted radiation will enter the feed directly and for a perfectly circular secondary reflector the feed will be on the caustic of the diffracted rays. Thus relatively large interference effects with the reflected rays can be expected. This can be significantly reduced by making the subreflector slightly un-round and/or serrated. The un-roundness need only be of the order of a wavelength; thus a "rough" edge with serrations of a few millimeter would also be effective.

All three precautions were taken in the design and construction of the 30-m millimeter telescope of IRAM. This telescope exhibits low baseline ripples as illustrated in Fig. 6.16. A small, fast residual ripple of 42 m equivalent length is still visible. It most likely originates in the electronics, but could be caused by a reflection path in the telescope of 42 m length.

0 0

Post a comment