The MOS/SIS instrument is attached to the Cassegrain bonnette (acquisition/guiding/rotator unit) at the f/8 focus of the CFH 3.6 m telescope. The image scale at this focus is 139.4µm per arcsecond. SIS and MOS are located on two opposite sides of the large, stiff, octagonal structure with 4 openings, or side ports, attached to this bonnette. The two remaining ports are presently unused.
The present optics of MOS were designed to work between 0.36µm and 1.00µm and were optimized to provide good images over this wavelength region and over the entire 10' x 10' useful field. The MOS optical scheme is shown in Morbey (1992, Applied Optics, 31, 2291). The 46mm pupil is 160mm from the last element of the f/8 collimator and is made to be coincident with the back side of the grism. The back focal distance is 37mm from the final lens element of the camera, and the exit pupil is -175mm from the final f/2.75 focus. The final scale is 47.7µ per arcsecon, or 0.3145"/pix for CCDs with 15µm pixels.
With this optical design, all rays, coming either from the edge or from the center of the field are constrained to be coincident at the pupil and rays from each object point are parallel as they pass through the grism so that the same region of the grism disperses the light from any point of the field in the same way.
All of the collimator lenses and most of the camera lenses (including two aspheric) were fabricated by Applied Physics Specialties of Toronto. PSK3 and FK54 glasses are used for all optical elements except the grisms.
Spot diagrams for a variety of wavelengths and positions are given in Morbey (1992, Applied Optics, 31, 2291). Spots originating from a 9 x 9 grid of stars spread over a 10' x 10' field are shown as they would appear to a detector in Crampton et al. (1992, proceedings of the ESO conference "Progress in Telescopes and Their Instrumentation", M.H. Ulrich Ed.). The images remain good over the entire field, and the geometric distortions are quite small, amounting to a maximum of 0.95% at the edge of the field. With the mask-holder and grism in place, the useful field is a little less than in imaging mode (see remarks in "Mask Preparation" in Chapter 3 and "Preparing a Mask File" in Chapter 5).
Considerable attention was paid to minimizing parasitic and scattered light. Focal reducers of this type often suffer from light scattered backwards from the CCD chip into the camera where it is reflected back to the detector in the form of a diffuse spot 5 - 10% above the background in the center of the image. Cures include providing a sufficient distance between the last optical element and the chip, as well as efficient anti-reflection coatings. MOS is clearly quite good in this regard, with a background concentration of <0.5%.
Ghost images of bright stars are also formed by reflections from the last surface of the last camera lens. In MOS, the surface brightness of the central ghost image is about 0.5% of that which is contained within a 1" image on the detector. Observers could have the feeling that these ghosts are strong but they are, in fact, very seldom a problem in spectroscopy. For direct imaging in fields with very bright stars, taking two exposures with a slight offset can help avoid the loss of faint structures. We might also suggest the use of other direct imaging instruments if precise photometric measurements are required in such "difficult" fields.
The image quality was ultimately measured on stars observed with MOS in imaging mode with broad band filters. The instrinsic image quality of the MOS optical train is best described in Figure 2: this shows that the MOS optical train will produce images with 25µm FWHM at the center and 45µm FWHM at the 5' radius. The reason for this difference is that the MOS detector focal plane is curved, so that the best focus cannot be achieved simultaneously at the center and edges of the field. The effect amounts to a defocusing of 45µm from center to edge. To balance the focusing over the full field, the MOS camera focusing should be optimized at 5mm (~ 2') from the optical axis. Then, assuming a quadratic combination, the FWHM as measured on stars on the CCDs will be on average for this balanced focusing. For an instrument focusing optimized at the center of the field, the fwhm at center will be and in the corners (with, in fact, an obvious elongation of the images). This means that, for the typical value FWHM(atm) = 0.7", the image quality would be at the center of the field and 1.1" in the corners. Thus, this option for focusing should be adopted only if you do not work with the full field of the instrument.
The main mechanical components in MOS that the observer can control are the following:
The measured flexures are plotted in Figure 4. The internal flexures of MOS for the telescope tracking a star are therefore expected to be a maximum of 15µm at the detector focal plane for a motion of 4h in hour angle or equivalently an average of 0.075"/hour.