All spectroscopic observations in this paper were obtained with the Coudé feed telescope at Kitt Peak National Observatory (KPNO) during five runs in March 1994, February-March 1995, January 1996, April 1997, and January 1998. The 1994, 1995 and 1998 data were obtained with a 8002 TI CCD (TI-5 chip, 15µ pixels) with grating A, camera 5, and the long collimator resulting in a resolving power of 38,000 at 6420 Å . For the 1996 and 1997 spectra we employed the larger 30001000 CCD (Ford F3KB chip, 15µ pixels) with an otherwise identical spectrograph set-up but centered at 6500 Å . The latter chip allows for a full 300-Å wavelength field but at the somewhat lower instrumental resolution of 32,000. Table 1 is a summary of the spectroscopic observations. Plots of representative spectra are shown later in Fig. 5.
All data were reduced with IRAF and consisted of bias subtraction, flat fielding and optimized aperture extraction. Frequent wavelength comparison spectra and spectra of bright radial-velocity standards were obtained at least once during each night to ensure an accurate wavelength calibration. Radial velocities of HD 51066 were then derived from nightly cross correlations with the IAU velocity standards 16 Vir (K0.5III, km s-1 ), Gem (K0III, km s-1 ) or Tau (K5III, km s-1 ) and IRAF's fxcor routine. The results are listed in Table 1 along with the errors from a Gaussian fit to the cross correlation peak ().
The analog-to-digital units in the continuum at 6500 Å correspond to signal-to-noise ratios of 200:1. Integration time was either 3000 or 3600 sec depending upon seeing conditions. Twenty flat-field exposures with a tungsten reference lamp were taken at the beginning of the night and again at the end of the night. These fourty flat fields were co-added and used to remove the pixel-to-pixel variations in the stellar spectra. Neither the TI CCD nor the F3KB CCD show obvious signs of fringing near 6400-6700 Å and no attempts were made to correct for it other than the standard flat-field division. Continuum fitting with a very low-order polynomial was sufficient to find a satisfactory continuum solution.
Throughout the years 1996 to 1998, continuous V(RI)c photometry has been obtained with Wolfgang-Amadeus, the 0.75m Vienna Observatory Twin Automatic Photoelectric Telescope at Fairborn Observatory in southern Arizona (WA-APT, Strassmeier et al. 1997b). BV data for the season 1995/96 were kindly made available by G. Henry at Tennessee State University prior to publication (Henry 1997). Altogether, 597 new VI measures for 1997/98, 118 VRI points for 1996/97, and 20 VRI and 92 BV measures for 1995/96 were obtained, each the mean of three readings of the variable and four readings of the comparison star. Johnson BV photometry prior to 1995 was taken from Henry et al. (1995b), whose data were gathered with the 0.4m Vanderbilt/TSU robotic telescope also at Fairborn Observatory. All V-band data are plotted in Fig. 1. Note that the WA-APT data were transformed to the Johnson-Cousins V(RI)c system and used HD 48840 (V=7:m 51, B-V=1:m 03, V-I=0:m 98; ESA 1997) and HD 45947 as the comparison and check star, respectively while the Vanderbilt/TSU APT data were transformed to the Johnson UBV system with the same comparison star but HD 50904 as the check star. The standard error of a nightly mean from the overall seasonal mean was for both telescopes 0:m 004 in B and V (0:m 0025 for the WA-APT since Oct. 1997) and 0:m 006 in Rc and Ic.
© European Southern Observatory (ESO) 1998
Online publication: July 20, 1998