Laboratory tests

Our spectrograph was first assembled and tested in the lab in Poznan Observatory cellar. The tests demonstrated proper collimation and focusing with no stray light between spectral orders (Fig. 5). This leaves open future option of recording two interleaved spectra at once, e.g. for calibration purposes. Arc lines at 6500 A yield instrumental profile of our spectrograph with FWHM =2 pixels i.e. 0.2 A or 8.5 km/s, corresponding to resolution R=33000 (Fig. 4). Thus resolution of our spectrograph is consistent with the original design.

Fig. 1. Subset of the flat field lamp spectrum. We use a standard wolfram lamp, producing a thermal emission. A significant curvature of orders can be easily noticed. The color is artificial.

Fig. 2. The perpendicular to dispersion cross section of Fig. 1. Sharp spikes as order cross section demonstrate proper colimation and focusing. The FWHM of a cross section is 3 pixels, consistent with that of the prototype Musicos. A horizontal axis on this plot is calibrated in pixels, the inter-order gap is 21 pixels.

Fig. 3. Enlarged fragment of the neon lamp spectrum (spectrum orders are horizontal). Inclination of the images of lines is produced by cross-dispersing prism, a natural consequence of the echelle design. The line marks the cross section presented Fig. 4.

Fig. 4. The instrumental profile in the direction of dispersion, as demonstrated by the cross section of the 6506 A line of Ne lamp, marked in Fig. 2. Its FWHM=2 pixels corresponds to 0.2 Angstroms or 8.5 km/s. This yields a resolving power R=33000.

Fig. 5. Fragment of the daylight solar spectrum obtained by directly feeding the sun rays into the fiber. In Figs. 6, 7 we compare a marked fragment of this well exposed, high S/N spectrum with solar atlas around H alpha.

Fig. 6. A cross section through the Balmer H alpha line from our solar spectrum (Fig. 5). The plot is about 4 A wide.

Fig. 7. The fragment of the solar atlas spectrum, scaled to match Fig. 6. The wavelength is given in nanometers. Notice the double line at the left, with components separated by 0.15 A unresolved in our spectrum, having the instrumental FWHM of 0.2 A.