next up previous contents index
Next: Fiber feed and preslit Up: CAOS - The new Previous: Design requirements   Contents   Index

Optical Design

In order to meet design requirements we selected a bench-mounted, fiber fed, cross-dispersed echelle spectrograph. As known, resolution in Littrow configuration is given by the resolution-slit product:

\begin{displaymath}
R \varphi = 2 (d/D) \tan\delta
\end{displaymath} (4)

where $R$ is the resolution, $\varphi$ is the angular slit width on the sky, $d$ is the beam diameter at the collimator, $D$ is the telescope diameter and $\delta$ is the blaze echelle angle. Overall spectrograph costs generally increases with beam size, so we want to minimize beam size. As trade-off, we selected a 10 cm collimated beam dispersed by a R4 echelle grating, focused with an optimesed all-dioptric camera onto a 2K$\times$2K, 13.5 $\mu$m pixels, CCD.

Unvignetted resolution defined by fiber projection is about 28000, while higher resolution modes will be available adding an entrance slit between preslit optics and spectrograph collimator. The maximum spectral resolution is $R=63000$ with a 2 pixel sampling. Interorder separation allows us to accomodate two fibers projection for spectropolarimetric studies. The spectrograph will be installed onto an optical bench in a thermally and mechanically isolated environment inside a room in the telescope pillar, allowing a stable spectrum position.

Figure 2.1: Optical layout of the spectrograph. Light enters from fibers through a preslit system into the spectrograph entrance slit.
\includegraphics[width=14cm]{CAOS/optlayout.eps}
This spectrograph is a white-pupil echelle spectrograph (e.g., like UVES at VLT and SARG at TNG (figure 2.1).


next up previous contents index
Next: Fiber feed and preslit Up: CAOS - The new Previous: Design requirements   Contents   Index
Innocenza Busa' 2005-11-14