Since October 1999 inclinometers are used at the 30m telescope. They measure two parameters of the pointing model (P4 and P5) that reflect the tilt of the azimuth axis with respect to the astronomical zenith. P4 and P5 change considerably in time, with daily variations of up to , due to thermal deformations of the concrete pedestal supporting the azimuth bearing. P4 and P5 are being measured every time the telescope moves 50 degrees or more in azimuth, with a precision of better than .
Since January 2000 P4 and P5 have been updated regularly by the AoD as soon as a significant change has been noted. We have compared the average scatter of the pointing offsets for a period of 21 months before October 1999 and for a period of 16 months after January 2000. This average scatter was measured by the scatter of the observations for a pointing model before implementing the new model. It represents the state of the pointing model after, on average, 8 days (after January 2000), respectively 12 days (before October 1999).
The precision of the pointing model has clearly improved after January 2000 due to the use of the inclinometers. Whereas the scatter of the pointing offsets was arcsec before October 1999, it is now on average arcsec.
In the near future, the latest values of P4 and P5 will be updated automatically at every pointing measurement.
In early May 2001 the IRAM 230 GHz SIS HEterodyne Array Receiver (HERA) has been successfully installed in its intermediate configuration (single linear polarization, 9 elements) at the 30m telescope. The multi-beam receiver currently undergoes commissioning. HERA is part of the new 30m instrumentation and was included in the new cabin layout from the very beginning. HERA significantly improves the speed for raster and on-the-fly mapping of spectral line emission in the range from 210 to 276 GHz, and will eventually also allow for new observing modes like dual beam wobbler switching or imaging polarimetry.
The receiver with its electronic, optical and cryogenic components has been integrated into a compact mechanical support structure. Its main features include a closed cycle cryostat with modular RF building blocks, a cold internal optic with a single circular cryostat window and a compact K-mirror field de-rotator in front of the cryostat. For more technical information see Schuster et al. (1999) 1. Due to its particular optic and position in the optical path of the 30m telescope HERA will not be available simultaneously with other receivers, but switching is possible within minutes.
The tuning range of HERA is 210-276 GHz in single (SSB) or double (DSB) sideband mode with an IF bandwidth of 1 GHz. Tuning is automatic but temporarily restricted to a limited set of frequencies (see the ``Call for Proposals'' later in this Newsletter). Receiver temperatures increase from 110 K T at 210 GHz to 380 K at 270 GHz, equalling the performance of the single-beam receivers. The de-rotator allows to optimize the array orientation for raster mapping and efficient sweeping during OTF observations with high precision. A particular pointing model is not required for HERA.
Figure 1a shows the array pattern on the sky obtained in continuum (1 GHz at 230 GHz) observation and measured with a high signal to noise OTF sweep over Saturn under medium weather conditions. The right element of the center row shows some irregularities due to microphonics of the corresponding mixer at this frequency. The beams have a spacing corresponding to 2 beam widths. Due to limited pointing accuracy during the observations the whole array pattern is shifted about to the north. The restored image of Saturn (Fig. 1b) shows the characteristic deviation from circular symmetry due to the oblation of Saturn itself and emission from the rings.
HERA's second polarization module is currently in preparation. It will have an independent frequency setting and its 9 elements will spatially coincide with those of the 1st polarization. The data acquisition rate of HERA in its dual polarization configuration will be considerable, and a particular effort will have to be made in terms of computer hardware and data reduction software.
Since some time we are studying the telescope's thermal life in order to eventually improve its focus and pointing. For this purpose we have installed 150 temperature sensors, i.e. 105 in the reflector back structure (BUS), and 45 in the box-shaped yoke. The accuracy of the readings is approx. C; the data are recorded every 5 minutes. These temperature measurements are used in a finite element model (FEM) to predict the temperature-induced mechanical deformations of the BUS, the yoke, and the subreflector support legs. We are able to use these data in the prediction of the focus and the beam shape; we are not yet fully successful in the prediction of the pointing.
The attached figures give some illustration of the potential of these measurements.
During the comissioning of the 9-channel 1.3 mm (230 GHz) HERA receiver we were confronted to determine the simultaneous focus from a focused (F = -2.43 mm), an in-focus ( F = - 1.73 mm), and an out-of-focus (F = -3.13 mm) 1GHz-BW continuum map of Uranus (diameter ). The de-focusing was produced by a displacement of the subreflector of mm, i.e. approx 0.5 lambda. Figure 2a shows the focused map of Uranus, Fig. 2b the out-of-focus map. The out-of-focus map shows the same astigmatic beam pattern, for all channels.
It is known since some time that the 30m telescope has a time-variable residual thermal deformation which appears primarily as astigmatism with principle axes in the direction up-down (EL) and left-right (AZ). This astigmatism is produced by temperature-induced main reflector surface deformations. The astigmatism during the measurements shown in Fig. 2a-b had an amplitude of 0.08 to 0.09 mm, while most of the time the astigmatism is smaller. This amplitude is determined from the analysis of Fig. 2a-b, and is also obtained from the temperature measurements and the FE calculation, shown in Fig. 3a. The corresponding predicted de-focused beam pattern is shown in Fig. 3b. There exists excellent agreement between the measurements and the prediction from the FEM and beam pattern calculation.
However, the observer at the 30m telescope does not need to worry. Observations are never made with a beam de-focused as shown in Fig. 2b; the focused beam (even with astigmatism) is circular and only broader than the perfect beam; the loss in mainbeam sensitivity is below 5% at 230 GHz. Often the amplitude of the astigmatism is smaller than the value shown in this illustration.
Finally we note that the 30m telescope works already significantly better than the original specifications (for instance the surface accuracy). Further improvements require a detailed understanding of time variable processes, where the FE model - as shown above - can make valuable contributions.