2) The antenna tolerance theory of several independent large-scale and small-scale wavefront (reflector surface) deformations provides a consistent description of the 30-m telescope beam. This theory makes use of the concept of individual correlation lengths (L i) and individual rms-values ( i) of the independent error distributions [ i]. An estimate of the correlation lengths can be obtained from details of the reflector surface construction, in particular for a design of (mini-)panels and panel frames.
3) We find from the analysis of the 30-m reflector that the Gaussian correlation length distribution is a good representation of the actual situation. The observed wavelength-scaling of the error beams gives confidence in the correctness of the theoretical predictions and the empirical values.
4) The effective rms-value of the wavefront (reflector surface) is the root-square-sum of the individual rms-values i. As expected, this value agrees for the 30-m reflector with the effective rms-value obtained from holography data and efficiency measurements. Compared with the errors of efficiency measurements, the standard Ruze relation (for L i = 0) can be applied on the 30-m telescope without loss of precision.
5) Any improvement of a paneled reflector surface with several error distributions should act, if possible, on all error components which, however, dependent of the reflector construction, may not always be possible. It is important to consider also the correction of large-scale wavefront deformations (for instance by using adaptive optics, Greve et al. 1996) which deform the central area of the beam and which may contain a significant amount of the total power.
6) Any further improvement of the 30-m reflector surface can, at the present state, only be made by further improvement of the panel frame adjustment. In this case, the parameters of the improved beam are obtained in the way as outlined in Sect. 4, supported by Moon scan measurements.
Copyright The European Southern Observatory (ESO)