The Effelsberg 100m radio telescope

Radio astronomy in Germany was established at the University of Bonn with the construction of a 25-m telescope in 1957. In 1962 Otto Hachenberg was appointed professor of radio astronomy. He came from the Heinrich-Hertz-Institute in Berlin, where, together with industry, he had built a 32-m transit instrument. Lacking finite element analysis and computing power, they had studied experimentally how spaceframe structures deform under gravity. Here an indication of homologous deformation became apparent without the analytical foundation, provided later by von Hoerner. In Bonn, Hachenberg began to plan for a giant fully steerable telescope and a proposal for financing was presented to the Volkswagen-Foundation in 1964. The reaction was positive under the condition that the capability of an effective operation of the instrument could be demonstrated. This led to the establishment of the MaxPlanck-Institute for Radio Astronomy in Bonn in 1965, of which Hachenberg became the founding director.

Design studies for a 80-m diameter antenna were made by Krupp and MAN, where the ideas of homology, as experimentally experienced by Hachenberg and theoretically worked out by von Hoerner, were applied. Eventually this led to a joint venture of the two companies in 1965 with the task of designing and building a fully steer-able telescope of 100 m diameter suitable for observation at a smallest wavelength of

2-3 cm. This requires a reflector accuracy of mm and a pointing accuracy of <10 arcseconds, both a significant improvement over existing large telescopes, all of which were much smaller than 100 m (Hachenberg, 1968, Geldmacher, 1970, Altmann, 1972).

Suspending Radio Telescope

Fig. 7.7. Exploded view of the three main sections of the antenna. The upper section is the BUS, supporting the panel units through a space frame structure. The middle octahedron supports the BUS at points B and is itself supported on the alidade at points A. The upper half of the octahedron forms the quadripod support of the subreflector. (Drawing Krupp)

Fig. 7.7. Exploded view of the three main sections of the antenna. The upper section is the BUS, supporting the panel units through a space frame structure. The middle octahedron supports the BUS at points B and is itself supported on the alidade at points A. The upper half of the octahedron forms the quadripod support of the subreflector. (Drawing Krupp)

The reflector backup structure (BUS) was conceived as a spaceframe with a high degree of rotational symmetry. This was guided not only by the idea of homology, but also by the still limited computational power of contemporary computers. For the structural analysis of the BUS only a pie-shaped section of one-twenty-fourth needed to be considered. The engineers achieved the desired result with a method of "trial and success", combining their practical experience with the new ideas of homologous structures. The BUS is supported from the back by an umbrella-type cone of spokes (Fig. 7.7). This structure (blue) is attached to the elevation structure (red) at only two points B: the tip of the umbrella and the center of the spoke wheel in the lower plane of the BUS proper. This two-point suspension provides a stiffness which is symmetrical with respect to rotation. This in turn assures a homologous behaviour in which the

Radio Telescope Cross Section

Fig. 7.8. Cross-section through the Effelsberg 100-m telescope. The reflector support structure (blue) is highly symmetrical and is connected to the elevation octahedron (red) at two points only. The octahedron is supported at two elevation bearings by the alidade (black), which runs on a 64 m diameter railtrack. (Drawing Krupp)

Fig. 7.8. Cross-section through the Effelsberg 100-m telescope. The reflector support structure (blue) is highly symmetrical and is connected to the elevation octahedron (red) at two points only. The octahedron is supported at two elevation bearings by the alidade (black), which runs on a 64 m diameter railtrack. (Drawing Krupp)

elastic deformations under varying elevation angle will result in a surface close to a paraboloid. The elevation structure (red) is essentially an octahedron, as suggested by von Hoerner (1967) in his classic paper. The quadripod support for the primary focus cabin / subreflector forms one half of the octahedron. The lower half is composed of the elevation bull-gear and two beams to the elevation bearings, where the entire structure is supported at points A by the alidade (black). This solution assures that the quadripod does not have any influence on the reflector. Also there is no physical connection between the octahedron near the elevation bearings and the BUS. Contrary to most earlier designs the surface is not subjected to the "point" loads of the quadripod and the support at the elevation bearings.

Fig. 7.9. The 100-m diameter Effelsberg telescope of the MPIfR, Bonn, Germany. The "umbrella" reflector support and the rear half of the octahedron structure are well visible. The cabins at the elevation bearings are 50 m above the ground. The collar around the edge of the reflector serves to reduce pick-up of ground radiation. (N. Junkes, MPIfR)

Fig. 7.9. The 100-m diameter Effelsberg telescope of the MPIfR, Bonn, Germany. The "umbrella" reflector support and the rear half of the octahedron structure are well visible. The cabins at the elevation bearings are 50 m above the ground. The collar around the edge of the reflector serves to reduce pick-up of ground radiation. (N. Junkes, MPIfR)

The alidade runs on a railtrack of 64 m diameter through 4 bogies with 4 wheels each. The elevation drive originally contained two dual anti-backlash drives, but instabilities in the system forced the removal of one drive, fortunately without ever impeding the operation of the telescope. A cross-section through the telescope is shown in Fig 7.8, where the same colours have been used to show the major sections.

The optics of the Effelsberg telescope is somewhat unusual. First, it employs a Gregorian two-reflector system with an elliptical secondary mirror of 6.5 m diameter. The primary focus is accessible through a hole in the secondary reflector to accommodate long wavelength (>20 cm) feeds. Second, the focal ratio of the primary reflector is 0.3, which makes the dish quite deep. To minimise spill-over radiation from the ground, a vertical shroud extends from the perimeter of the reflector. This was especially useful, because the telescope is located in a narrow, deep valley (Fig. 7.9) to shield it from radio interference.

The surface of the reflector over the inner 65 m diameter consisted originally of aluminium sandwich panels with an rms accuracy of 0.25 mm. Similarly as with the WSRT the epoxy bonding between the surface plates and the honeycomb core delaminated after about ten years and all panels were replaced by aluminium plates reinforced with backing ribs (cassette panels). This type was used ab initio for the area between 65 and 80 m diameter. To decrease wind loading, the outer area was originally composed of stainless steel wiremesh with 6 mm mesh size. With the replacement of the sandwich panels, the mesh was also replaced with cassette panels, perforated with 7 mm diameter holes to decrease wind loading. Thus the telescope is not effective beyond the inner 80 m for wavelengths smaller than about 2-3 cm.

The performance of the instrument is quite impressive. After setting the surface with the new panels with the aid of satellite holography at a frequency of 12 GHz, the rms surface error is 0.45 mm at the setting angle of 32 degrees. It increases to only 0.7 mm at an elevation angle of 80 degrees. This increase in surface rms should be compared with the actual deformation of the structure, which is 76 mm at the edge of the dish. The pointing accuracy is better than 10 arcseconds with a repeatability over time scales of one hour of 2 arcseconds (Hachenberg et al., 1973). The telescope is routinely used at 3.5 mm for VLBI observations. The only active control necessary for achieving this performance is the adjustment of the subreflector (or primary focus feed) to the focus of the best-fit paraboloid. During the commissioning of the telescope a perfect match was found between the prediction based on structural calculation and the measured adjustment. This is a good demonstration of the high quality of the structural design. Over the range of elevation angle from 15 to 85 degrees the lateral shift in the focus position is about 120 mm, while the axial shift is 15 mm.

Recently, another telescope of comparable size and performance has come into operation; the Green Bank Telescope (GBT) of NRAO (see Ch. 1). Its offset reflector is slightly larger (110 x 100 m) for "political" as well as technical reasons, as explained in Chapter 6.3.2. The large surface elements of the GBT are adjustable through motor-controlled actuators, which enables a constant rms surface error over the entire elevation range, if the necessary corrections are known. The current level of finite element analysis makes this a viable option. This challenge to the Effelsberg telescope is being met by the installation of an active subreflector of 6.5 m diameter, designed and built by MT-Aerospace. It consists of 96 individually adjustable, accurate (10 mm rms) aluminium facets with an overall surface error of less than 80 mm. Because the gravitational and slowly varying thermal deformations of the primary reflector have a rather large scale-length and can be computed reliably with a finite element analysis, the adjustment of the relatively small number of subreflec-tor facets will restore the full accuracy of the dual-reflector Gregorian system. Undoubtedly, the Effelsberg telescope is a splendid piece of engineering, beautifully executed and an impressive example of the power of homologous design.

Was this article helpful?

0 0

Post a comment