3. Observations and data reduction
The observations were done with the IRAM 30-m telescope near Granada, Spain between 10th Feb. and 15th Feb., 1998. The receiver we used was the 37-channel bolometer described by Kreysa et al. (1998), with feeds arranged in a hexagon around the central feed, and beam separations of 20". The sensitivity of each channel was 70 mJy/s. The beam size was as derived from pointing scans. All our sources were point-like to this beam. Opacity measurements were made via skydips, from which we derived zenith opacities of 0.1-0.8. Despite the high atmospheric attenuation of the signal, the observations (generally done at high source elevations) could be performed satisfactorily, owing to the relatively quiet atmosphere present during our observing run (resulting in a relatively low sky noise). The calibration factor to convert the observed counts into flux densities was determined by observing the source K3-50A, which has a total flux density of 10 Jy and a peak flux density of 7.6 Jy for the beam size of our measurements. The measures were finally corrected for atmospheric attenuation and the elevation-dependent gain of the telescope. The pointing was frequently checked by cross-scanning the sources 0923+392, 1144+402, 1308+326, and 1418+546. The pointing accuracy was found to be . The same sources were used to adjust the focus at regular intervals.
The sources were observed in an ON-OFF mode in which the wobbling secondary mirror of the telescope moved at a frequency of 2 Hz between the source position and positions located at in azimuth. Each source was observed between 3 and 5 times, with 10 ON-OFF pairs each. The differences of the flux densities between the main and the reference beams were averaged by weighting inversely proportional to the square of the r.m.s. noise of each measurement. The resulting integration time spent on each source was thus typically 50 minutes, and the final r.m.s. noise of each measurement was between 2 and 5 mJy. In addition, the calibration may be uncertain by 15% owing to day-to-day variations in the system gain.
The data were recorded with all receivers, with the central one pointing at the source. As the angular extents of the target sources are in the range of a few arcseconds, the outer channels contain only sky emission. Hence, they were used to subtract a mean sky level from the central channel. Using this procedure, we assumed that the scale size of the fluctuations of the sky emission is large compared to the total coverage of the bolometer array (), which corresponds to 1.5 m at a distance of 5 km. Together with the wobbling rate, this ensures efficient rejection of sky noise, since the time scale of atmospheric fluctuations is much larger (see e.g. Altenhoff et al. 1987).
In Table 1 we list the measured flux densities and 1- errors, inclusive of noise and calibration uncertainty, at 230 GHz.
Table 1. Flux densities at 230 GHz
© European Southern Observatory (ESO) 1999
Online publication: April 28, 1999