3 Observations and data reduction

The observations were carried out with the 6 m telescope of the Special Astrophysical Observatory of Russian Academy of Sciences (SAO RAS) in the period from December 1992 to October 1996. Medium resolution spectrograph SP-124 in the Nasmyth-1 focus was used over the whole period with two different modes of registration. For the period prior to December 1995 the spectra were obtained in the IPCS mode (see the description of IPCS in Drabek et al. 1986; Afanasiev et al. 1986). Since January 1996 we used for registration Tektronix CCD 1024$\times$1024 provided by Astrophysical Institute Potsdam (AIP, Germany) in the frame of mutual cooperation between AIP and SAO RAS.

3.1 IPCS observations and data reduction

Spectra obtained in the IPCS mode with the grating B1 (600 grooves/mm) cover the range of 3600-5500 Å with a scale along the dispersion 1.9 Å/pixel. The spectra were obtained through round apertures with a diameter of 1.5'' to 3'' depending on the seeing, centered on the brightest part of the galaxy, usually near its center. Sky spectrum was accumulated simultaneously in the aperture of the same size displaced by 20'' along the current direction to zenith. Typical exposure time was 3-5 min (snap-shot mode) in order to detect the strongest emission lines [OIII]$\lambda\lambda$4959, 5007, H$\beta$and [OII]$\lambda$3727 from the spectra with low S/N ratio, usually not exceeding 5-10 in continuum.

During photometric nights we have observed also spectrophotometric standards from the KPNO list (Massey et al. 1988). These observations then were used to correct the spectra of the program galaxies for the system spectral response. He-Ne-Ar source was used as the reference spectrum. Flat field was accumulated every night before and after the observations and all the spectra were corrected for the flat field after azimuth and modulation correction of the original 1024-channel arrays. The data reduction was done in MIDAS (93NOV version), with several programs from the context Spec combined into the special package and adopted to perfom automatic mode reduction for IPCS 1-D spectra (Kniazev 1994).

3.2 CCD observations and data reduction

The observations in 1996 were carried out with a new CCD-detector in two different set-ups. In February 1996 we used B0 grating (300 grooves/mm) with the wavelength coverage of 3600-7000 Å and a scale along the dispersion 5.2 Å/channel and B1 grating (600 grooves/mm) with the interval 3600-5400 Å and a scale along the dispersion 2.6 Å/channel. For an experimental set-up of the CCD-detector there was a significant vignetting which limited the actual wavelength range for both gratings. The spectra were obtained through a long slit of 1''-2'' width and 45'' length. For the October 1996 run the CCD set-up allowed the coverage of the range 3600-6000 Å with a scale along the dispersion 2.3 Å/channel and B1 grating. In these observations we used the registration system NICE under MIDAS (Kniazev & Shergin 1995). As in the case with IPCS observing mode the exposure time for both set-ups did not usually exceed 2-3 min which was enough to recognize the strongest emission lines from the low S/N ( 10) spectra.

Ten exposures have been obtained both in the evening and morning twilight to get the estimate of CCD detector's flat field. All primary reduction of CCD spectra was conducted in the standard way, using MIDAS context Long (94NOV version). 2-D CCD images after the correction for the dark noise, debiasing and sky-subtraction were transformed to linear wavelength scale and converted to 1-D spectra with the following correction for the atmospheric extinction and standard flux calibration. The same as for the IPCS-mode He-Ne-Ar source was used to build up the dispersion curve. Spectrophotometric standards have been observed at least twice during the night to allow the correction for the spectral response of the system.

The information on the set-up of the spectral equipment, used wavelength range, spectral resolution and the number of the observed objects for each run is shown in Table 1.

 

Table 1:   Journal of observations