MOSCA is a multi-purpose instrument which allows
  • direct imaging using
    • filters
    • a Fabry-Perot Interferometer as tunable narrow-band filter
  • long-slit spectroscopy of faint objects at low to moderate resolution (<1000)
  • multi-object spectroscopy using user-supplied masks
Click here for manual and for latest news!


MOSCA is a focal reducer, installed at the Richey-Chretien Focus of the 3.5m telescope on Calar Alto. The reduction ratio of the optical system is 3.7, i.e. the effective focal ratio is f/2.7. This gives an image scale of 3 pixel per arcsec and a total FOV of 11x11 arcmin (15 micron pixels). The imaging quality: 80% of the energy is contained within 15 micron radius, 100% within 30 micron radius over the spectral range 3300 to 10000 Angstroem. Various analysers are or will eventually be available: filters, grisms, FPI, and may be also polarizers. A thinned CCD with 2048x4096 15 micron pixels is used as detector. The data are stored on disk, and can be viewed and analysed with standard software packages (MIDAS, IRAF, IDL,...). The instrument is controlled with a user friendly GUI.


schematical drawing

MOSCA at the telescope


The calibration unit has 1 tungsten lamp as continuum source for testing purposes and 3 spectral lamps for wavelength calibration.


The aperture unit allows to select between:
  • 1 free position for direct imaging
  • 1 adjustable long slit with decker mask for spectroscopy. The slit can be viewed with the TV guider.
  • 2 positions for masks which are easily and quickly exchangeable (e.g. user supplied masks for multi object spectroscopy or strip masks for polarimetry)
  • 1 hole, used as optical reference point


Two filter wheels with 10 positions each are available and may be used simultaneously. The instrument is equipped with standard broad-band filters (U, B, BV, V, R, I, z).

Custom filters can of course also be used. Mechanically, the filters must have 70 mm diameter; several adapters are available for 50mm filters. Optically, the required diameter is 70 mmm for filterwheel I and 50mm for filterwheel II; use of smaller filters will cause loss of light in filterwheel I and loss of field in filterwheel II.


A set of 7 grisms with dispersions from 3 to 12 Angstroem per arcsec is available:

grism            lambda   lambda_central  dispersion
                  [nm]        [nm]        [nm/arcsec]
green_250        330-1000     570             1.2  
red_500          540-1000     850             0.6
blue_500         330-650      480             0.6
green_500        430-820      550             0.6
blue_1000        330-510      400             0.3
green_1000       420-650      530             0.3
red_1000         570-820      680             0.3

Efficiency curves are given in the manual .


A Fabry-Perot interferometer whith piezoelectrically controllable spacing can be used as tunable filter at a spectral resolution of 15-20 Angstroems over the wavelength range 5500 to 9500 Angstroems. Order selection is done with filters in filter wheeel II. It is very easy to use, and there is an automatic procedure to determine the wavelength calibration. If you plan to use the FPI make sure that a suitable order separation filter is available!


Since there is currently little interest to use polarizers, the realisation of polarizing optics has been postponed to an undetermined date.


Multi object spectroscopy is possible using user supplied masks which are put into the aperture unit. The masks have to be prepared by the observer before the observing run; there is no possibility to make masks on Calar Alto.

We have developed a procedure to design the masks and to do the critical alignment of sky and mask on the telescope. We offer all users to benefit from our experience and to make the masks according to our recipee in our mechanical shop. For more details see the manual .


  • first light: September 5-8 1996
  • installation of FPI: January 1998
  • MOS: 1998
  • Polarizing optics: tbd
  • new CCD (SITE 16a) April 1999
  • new grism red500 May 2000

    latest news

    15 May 2000: the red500 grism has been installed. Since the space in the grism wheel is limited, the focussing prism had to be taken out. The wavelength calibration looks very nice, see manual. The grism could not be tested on the telescope, so the efficiency is not yet known.

    15 May 2000: A 2x4k CCD with 15 micron pixels from EEV has been tested. Advantages of this CCD are fast read-out and good cosmetics. Disadvantages are lower QE than the standard SITE16a chip and very strong fringing in the red!

    March 2000: the V-filter is broken. Obviously the cementing of the 2 glass disks is peeling away. It will take several weeks to get a new one!

    March 2000: the red500 grism (peak efficiency at 850 nm, dispersion of 0.6nm/arcsec) is here and will be installed before the observing runs in May

    August 1999: A new grism has been ordered which has its peak efficiency at 850 nm and a dispersion of 0.6nm/arcsec; this grism will extend the spectroscopic capabilities of MOSCA to the red.

    April 1999: The LORAL 11i detector has been replaced by a better CCD which has much less fringing in the near IR which was a nuisance with the LORAL chip. The new CCD is a 2x4 k CCD with 15 micron pixels; this CCD has 4 bad columns but otherwise is fine

    the MOSCA team:

  • construction + mechanics: Bellemann, Benesch, Franke, Muench
  • electronics: Salm, Grimm
  • CCD: Marien
  • software: Zimmermann, Bruege
  • tests: von Kuhlmann
  • PI: Fried

    send questions or fan mail to