Astron. Astrophys. 323, 909-922 (1997)
2. Observations
This is the first paper that is based on observations with the
fiber optics cassegrain échelle spectrograph F OCES (Pfeiffer
et al. 1997) installed recently at the Calar Alto Observatory in
Spain. In short, this device provides a spectral coverage of more than
3000Å ( 70 orders) at a resolving power of
up to 40000, if equipped with a 10242
CCD ( 68mÅ/pixel at
5500Å). It produces almost no stray-light due to its special
white pupil configuration. In addition the user benefits very
much from a well defined blaze function that leaves little room to
personal interaction when determining the continuum, as is illustrated
in Figs. 1 and 2 for the rectification of the H
and H region.
![[FIGURE]](img10.gif) |
Fig. 1. Rectification of Moon (=reflected sunlight) spectra obtained with the FOCES spectrograph in the region of the H line: échelle orders #88 and #85 on the short- and long-wavelength side of the Balmer line show rectification curves that differ by less than 1% (top panel). The continua of the échelle orders 86 and 87 are found by interpolation
|
![[FIGURE]](img13.gif) |
Fig. 2. Same as Fig. 1, but for H : échelle orders #119 on the blue side and #114 on the red side of H are used for interpolation
|
The spectra we discuss below have been obtained on the third and
last test run in September 1995 at the 2.2m telescope of the Calar
Alto Observatory. Although the seeing was not excellent (1.5 to
) and the spectrograph not yet in the final
optimized stage, we obtained typical signal-to-noise values of
at 5000Å for a 10th magnitude star within
30 minutes. Table 1 gives a short log of the observations
relevant for the next sections. Except for HD 140283, all stars
were observed at least twice, with each exposure ranging from 3900 to
6900Å. Signal-to-noise ratios are calculated from Poisson
statistics, with a negligible CCD readout contribution. The actual
data however show an unexpected residual noise level of 0.5% as a
result of moving the fibre position with the telescope. For this
reason the true S/N values are not better than
200. The data reduction follows the common path
of e.g. Horne (1986). Details with respect to the F OCES spectrograph
and the data reduction will be given in due course (Pfeiffer et al.
1997).
![[TABLE]](img20.gif)
Table 1. Log of observations taken with the F OCES échelle spectrograph in September 1995. All spectra cover the range 3900-6900Å. Exposure times are integrated values in seconds. The nominal signal-to-noise ratio is valid around 5500Å
Fig. 3 shows a comparison of the FOCES Moon spectrum at the
red wing of H to the Kitt Peak Solar Flux Atlas
(Kurucz et al. 1984, dotted curve, here: renormalized by 0.5% "up"),
which serves as our primary reference source. Convolution with a
Gaussian of km s-1 results in
the (offset) dot-dashed curve, which clearly reproduces the FOCES data
to a high degree.
![[FIGURE]](img22.gif) |
Fig. 3. Comparison of the FOCES Moon spectrum to the Kitt Peak Solar Flux Atlas (dotted curve, renormalized by 0.5% "up"). A Gaussian of km s-1 applied to the latter results in the (offset) dot-dashed curve, demonstrating the great similarity of both spectra
|
© European Southern Observatory (ESO) 1997
Online publication: May 26, 1998
helpdesk.link@springer.de  |