2 Observations and data analysis

For the chemically peculiar stars listed in Table 1 echelle spectra were obtained in 1995 at the 2.1 m telescope of the Complejo Astronómico El Leoncito equipped with a Boller & Chivens cassegrain spectrograph and in 1997 at the 91 cm telescope of the Catania Astrophysical Observatory equipped with a Czerney-Turner echelle spectrograph.
The data were analysed the using IRAF package. The lines of the wavelength calibration lamp show that R=16000 for the 1997 data set and R=13000 for the 1995 data set. The achieved S/N was between 100 and 200. When possible, equivalent widths were measured by a Gaussian fit of spectral lines after a possible continuum slope; otherwise a measure of the area between the line profile and the continuum was obtained. Following Leone et al. (1995), we estimated the error in the measured equivalent width with the relation:  

$$\Delta W = \frac{1}{2} \left(2 \frac{v_{\rm e} \sin i}{c} \lambda\right) \frac{1}{S/N}$$ (1)

where the quantity in brackets is the total extension of the line as deduced from the rotational broadening. Adopted $v_{\rm e} \sin i$ values (Table1) are from SIMBAD with the exception of HD28843 whose projected rotational velocity was measured from the unblended SiII5865Å line.

The initial ephemeris of program stars were taken from Catalano & Renson (1984, 1988, 1997), and Catalano et al. (1991, 1993), and references therein. If necessary, periods were established using our spectral observations and Hipparcos photometry. A least squares fit of measured EW's and H$_{\rm p}$ magnitudes has been performed by adopting the function:

$$\begin{eqnarray} & & A_0 + A_1 \sin (2 \pi (t-t_0)/P + \phi_1) \nonumber \\ & & \qquad \qquad \qquad \qquad + A_2 \sin (4 \pi (t-t_0)/P + \phi_2) \end{eqnarray}$$ (2)

where t is the JD date, t₀ is the assumed initial epoch, P is the period in days. A sine wave and its first harmonic appear to be quite adequate functions to describe the light curves and the spectral variations (North 1984; Mathys & Manfroid 1985). The error in the period value has been evaluated according to the relation given in Horne & Baliunas (1986).

As to the coolest CP stars, the effective temperatures and gravities have been determined by means of ad hoc Napiwotzki et al. (1993) relations. As to helium peculiar stars, Hauck & North (1993) found that classical methods are still reliable to determine their effective temperature. Thus, we have used the Moon & Dworetsky (1985) grids as coded by Moon (1985). The source of Strömgren photometry was SIMBAD.

To ascertain if the selected stars present a peculiar helium abundance we have compared the measured equivalent widths of the HeI5876 Å line with the NLTE computations of Leone & Lanzafame (1998) for solar composition stars with $\log$g = 3.5, 4.0 and 4.5 and 9000 K $< T_{\rm eff} < 19000$ K.