|
Signal-to-noise ratios achievable for various types of star at
a range of selected blaze wavelengths and as a function of exposure time
are shown in these figures. The following conditions were assumed for
these plots:
- High-resolution image slicer.
- If the medium- or low-resolution
slicers are used, the efficiencies are increased by a factor of
1.2 or 1.25, respectively.
- Binning 1x1.
- Applying binning can improve the signal-to-noise
ratio for low count levels. To estimate this one must consider the CCD
read-out-noise of 3.4 electrons per unbinned pixel (rms)
in normal R/O mode, and take into
account the number of image pixels combined into one spectral pixel,
i.e. the width of the image slicer profile.
- Observation at blaze wavelengths where the efficiency is maximal.
- Seeing 1.0 arcsec.
- Due to the fixed aperture size of the CES fibre of
2.0 arcsec on the sky larger seeing values lead to a degradation of
the efficiency.
- Observation at airmass 1.0 (minimum extinction).
- La Silla mean extinction values were assumed.
The figures show the achievable S/N ratio for exposure times between
1 and 60 min for three different spectral types and two stellar
magnitudes. From top to bottom:
- B2 with V=6.0,
- B2 with V=9.0,
- A0 with V=6.0,
- A0 with V=9.0,
- G0 with V=6.0, and
- G0 with V=9.0.
As labelled in each figure the following blaze wavelengths
were selected:
- 3833 Å (cyan line),
- 4267 Å (blue line),
- 5026 Å (green line),
- 6462 Å (yellow line),
- 7798 Å (red line),
- 9046 Å (magenta line), and
- 9832 Å (black line).
In the accompanying figure that shows the total efficiency
(click here)
the positions of these blaze wavelength are marked by vertical
dashed red lines in order to show their location within the total
efficiency function.
| |
|
|
|