Follow-up spectroscopy of QSO candidates were carried out in several observing runs between 1986 and 1991 with the 3.5m and 2.2m telescopes at Calar Alto, Spain (Table 1 (click here)), partially as backup programs for times of poor weather. Non-photometric weather conditions affected however all observing epochs. As the development of the search techniques used only few fields, and the behaviour of the techniques was studied with emphasis on faint density spectra, most of the candidates observed had brightnesses close to the limit of the plates, e.g. 17.519.0. About half of the candidates were observed in only two fields located in the strip centered at and (Epoch 2000). In these fields plates of excellent quality were obtained and they were selected therefore as primary test fields.
Most spectra were obtained with the Boller & Chivens-Cassegrain spectrographs equipped with an RCA CCD (6241024 pixels with m pixel size) at a dispersion of 240Å/mm, covering a spectral range from 3800 to 6800-7200Å. On one single occasion (HS0843+2533) only the red spectral region beyond 6300Å was observed. The CCD was read out in binned mode (22 pixel), and the final resolution was 15Å. In 1987, February a Reticon covering the wavelength range from 3800 to 6100Å was used instead of the CCD detector. During two epochs the focal reducer with grisms and the RCA CCD was used, giving dispersions between 212 and 905Å/mm, and resolutions of . The spectral coverage was . Exposure times were mostly 15 minutes but ranged from 10 to 45 minutes.
|Nov. 29 - Dec. 1||1986||3.5 m||CS||240|
|Feb. 5 - Feb. 8||1987||3.5 m||CS||240|
|May 26 - May 29||1987||3.5 m||CS||240|
|Jan. 8 - Jan. 12||1988||3.5 m||CS||240|
|Jun. 6 - Jun. 11||1988||3.5 m||CS||240|
|Jan. 9-14, 19-22||1989||3.5 m||CS||240|
|Jun. 13 - Jun. 19||1989||2.2 m||CS||240|
|Dec. 13 - Dec. 21||1989||3.5 m||CS||240|
|Jan. 19 - Jan. 25||1990||2.2 m||CS||240|
|Jun. 21 - Jun. 24||1990||3.5 m||FR||290|
|Jul. 23 - Jul. 30||1990||2.2 m||CS||240|
|Oct. 16 - Oct. 23||1990||2.2 m||CS||240|
|Jun. 26 - Jul. 2||1991||3.5 m||FR||212/905|
The data reduction was carried out in Hamburg using the program package MIDAS. After standard bias- and flatfield-corrections the contribution of the night-sky was determined by fitting third order polynomials perpendicular to the dispersion for each pixel row in two windows outside the object spectrum. Cosmic ray hits were removed by median-filtering the rows, and the spectrum was extracted using an algorithm similar to the one described by Horne (1986). Wavelength scales were assigned by a linear fit to identified emission-lines in He-Ar comparison spectra. The flux calibration was done using observations of Feige34, EG247, BD+284211, HZ44 (Massey et al. 1988) and WD1736 (Greenstein 1984). For spectra taken during non-photometric weather conditions the flux calibration may have errors up to 40%. Corrections for atmospheric extinction or galactic reddening were not applied.