VIMOS MOS
To use the VIMOS Multi Object Spectroscopy mode users needs to prepare masks with the VIMOS Mask Preparation Software (VMMPS) that can be downloaded from here.
Grisms
Six grisms are available in MOS mode. The grisms are the same as used in IFU mode, but spectral resolution and wavelength coverage are different. The spectral coverage derived from the automatic reduction of spectrophotometric standard star observations with VIMOS pipeline is listed in the following table:
| Grism | Order sorting filter | Wavelength coverage | Spectral resolution (1" slit) |
Dispersion (Å/pix) |
Multiplex |
| LR blue | OS blue | 370-670 nm | 180 | 5.3 | 4 |
| LR red | OS red | 550-950 nm | 210 | 7.3 | 4 |
| MR | GG475 | 480-1000 nm | 580 | 2.5 | 1 |
| MR | OS red | 550-970 nm | 580 | 2.5 | 1 |
| HR blue | Free | 410-630 nm (1) | 2050 | 0.51 | 1 |
| HR orange | GG435 | 515-760 nm (2) | 2150 | 0.6 | 1 |
| HR red | GG475 | 650-875 nm (3) | 2500 | 0.6 | 1 |
(1) With HR blue grims the spectral range in quadrant 1(for a standard star) is 400-620 nm
(2) With HR orange grims the spectral range in quadrant 1 (for a standard star) is 505-745 nm
(3) With HR red grisms the spectral range in quadrant 1 (for a standard star) is 650-865 nm
IMPORTANT NOTE A new set of four holographic HR_red grisms has been installed in VIMOS on October 5 2005. Compared to the previous set, the efficiency has increased of about 50%. We remind that before that date, there was no HR_red grism in quadrant 4. The grism HR_orange was instead automatically inserted in quadrant 4 when requesting the grism HR_red. The efficiency curve of the old HR Red grisms is here.
MOS Spectral formats:
In Low Resolution, a spectrum will span 640 pixels along the dispersion direction, for both the blue and red grisms. This allow to stack 4 spectra along the dispersion direction (provided that in the field of view there are enough well spaced targets along the dispersion direction).
At medium resolution, a spectrum will span ~2000 pixels when used with the GG475 sorting order filter. It therefore possible to stack up to 2 spectra along the dispersion direction, provided that the slits are positioned at the edges of the imaging field of view.
At high resolution, a spectra will cover the whole detector length, i.e. 4096 pixels. The spectral coverage is therefore dependent of the position of the slit in the field of view. This is illustrated to the following table.
| Grism |
Slit position (1) in the FOV (arcmin) |
Wavelength coverage (nm) |
| LR blue | from -4 to +4 | 370-670 nm |
| LR red | from -4 to +4 | 550-950 nm |
| MR | from -4 to +4 | 500-1000 nm(2) |
| HR red | +4.0 | 705-950 nm |
| HR red | 0 | 630-870 nm |
| HR red | -4.0 | 565-800 nm |
| HR orange | +4.0 | 600-840 nm |
| HR orange | 0 | 520-760 nm |
| HR orange | -4.0 | 455-695 nm |
| HR blue | +4.0 | 480-690 nm |
| HR blue | 0 | 415-620 nm |
| HR blue | -4.0 | 365-565 nm |
(1) Slit positions are 0 (center) and -4.0, +4.0, corresponding to the edges
of the imaging field of view along the dispersion direction (8').
(2) the actual spectral coverage depends on the order sorting filter
Another constraints on the slit position is the presence of the zero and second grism diffraction orders. At low spectra resolution, every spectrum at first order will have a dimmed (a few %) second order spectrum, at twice higher spectral resolution, overlapping the adjacent spectrum next to it. This constraints the slit to be of the same height and aligned along the dispersion direction for a clean second order sky substraction.
MOS: Number of slits.
Apart from the geometrical considerations above, the number of slits that
can be accommodated in one mask obviously depends on the target density. Simulations
made by the consortium on real fields using the Mask Preparation Software
show that above a density of ~7E4 per square degree, it is possible to define
masks with up to 650 slits per field (4 quadrants) with slits ~10" long.
This number drops to 500 at a density of ~4E4 objects per square degree.
During commissioning up to 1000 spectra were taken on one masks. During ordinary operations users routinely create masks with more more than 100 slits (with low resolution grisms).
An example of 2 quadrants with about 100 slits in each mask is shown below
Instrument Flexures
Instrument flexures of different importance in the different quadrants strongly affect the data. To minimize their impact on the science, at a prize of increased overheads, MOS and IFU observations need to have attached a NightCalib template. The following page show the flexures as a function of rotator angle.