Astron. Astrophys. 319, 607-616 (1997)

2. Observations

The observations were carried out with the 30 m IRAM telescope on 14 January 1994 and June 1995. Most of the data present here were taken during the first run. Two SIS receivers were simultaneously used at 86 and 230 GHz. During this run, the sky was clear, the ambient temperature was close to - C and the relative humidity was nearly zero; consequently, the transmission at these two frequencies was excellent (a zenith opacity lower than 0.04 was measured at 230 GHz during the observations). The 3mm receiver was tuned in SSB mode at the frequency of the v =1 J =2 1 line of SiO; the attenuation of the image band being about 10 dB. The SSB receiver and system temperatures were 80 and 140 K respectively. The 1mm receiver was tuned, also in SSB mode, at the frequency of the J =2 1 line of CO, the SSB receiver and system temperatures being 100 and 180 K respectively. The main beam telescope efficiencies at the frequencies of the CO J=2-1 and SiO v=1 J=2-1 lines are 0.45 and 0.60 respectively (main beam brightness temperature for SiO can be translated into flux in Jy by multiplying the corresponding intensities by a factor 4.5).

The pointing was frequently checked by observing the SiO maser line itself using 16 channels of the 100 kHz filterbank. Both the absolute pointing of the 3mm receiver, and the alignment between the two receivers were better than . The focus adjustment was also checked every hour in the same pseudo-continuum mode.

The spectrometers consisted of two 512 1 MHz filterbanks, a 2048-channel autocorrelator, and a 256 100 kHz filterbank. In SiO the spectral resolution was 3.6 km s^-1 for the 1 MHz filters, and 0.07 km s^-1 for the correlator. In CO it was 1.3 and 0.05 km s^-1, respectively. For all stars, the autocorrelator spectra for SiO, when smoothed to 0.35 km s^-1, were similar to that produced by the 100 kHz filterbank. (i.e., no artificial broadening of the lines was observed in the autocorrelator). Due to the large intensity of the SiO maser emission, weak spurious features could appear after heterodyne mixing. In order to suppress such possible features, the sideband noise level of the phase-lock loop signal was adjusted to less than -40 dB. All observations were performed with the wobbling system with the OFF reference position away in azimuth from the source and with a wobbling time phase of 2 seconds. This observing procedure provides excellent baselines for spectral data. Line calibration was achieved using two absorbers at different temperatures and by observing the sky emissivity to derive atmospherical opacities. Fig. 1 shows a few SiO and CO spectra observed toward µ Cep taken with different spectrometers (a degree 1 baseline has been removed). The quality of the baseline allows a good comparison of the CO and SiO line wings in each observing run.

Fig. 1. CO J =2 1 and SiO v =1 J =2 1 emission toward µ Cep. All the spectra have been obtained in January 1994 except the bottom one which was taken in June 1995. From top to bottom the backends are : 1 MHz filterbank, autocorrelator with =320 kHz, autocorrelator with =40 kHz, 100 kHz filterbank and autocorrelator with =40 kHz. All the observations were obtained with the wobbling system. A degree 1 baseline has been removed for each spectrum. Note the important change in the SiO line profile between 1994 and 1995.

For the June 1995 run the observing procedure and instrumental configuration were similar to those of 1994 except for the pointing which was done on strong continuum sources. Hence, pointing accuracy in this second run was slightly poorer and around .

© European Southern Observatory (ESO) 1997

Online publication: July 3, 1998
helpdesk@link.springer.de