next up previous
Next: Procedure for Aligning the Up: Observing with Multislit Masks Previous: Observing with Multislit Masks

Preparatory Work in the Afternoon


You should write a tape at your home institution including the tabular output files from cosmicslitmask.f. Dump these files into your current COSMIC directory on the COSMIC Sparc computer (i.e. where the data are being written).

For each mask you need to create a file containing your alignment objects. This should be done in the afternoon before observing, and the file should be in the same directory that you are using for taking images. This can be done easily by editing the tabular output of the cosmicslitmask.f program: (1) remove the header line(s); (2) keep only 5-10 objects that are good for slitmask alignment; and (3) reorder them so that the objects go from left to right (in the plot of the slitmask). A good object for slitmask alignment is relatively bright and compact, so that an accurate centroid can be measured in a relatively short imaging exposure. Call this file ``mask.slits''.

You will need to mount your masks into holders and then insert them into COSMIC.

Take a direct image of each of the slit-masks you have just installed. This images should be taken with the FILTER=CLEAR, GRISM=CLEAR, and APERTURE=multislit of your choice. You can take these with the mirror closed, using only the ambient light in the dome (before 4PM), with an exposure time of 10s.

Display the image of the desired slit-mask (with the [*,-*] orientation). Once you have taken the images, use the IRAF script ``masksetup'' to mark the positions of the slits corresponding to the objects in the ``mask.slits'' file. In IRAF, type ``masksetup''. You will be asked for an output filename (typically ``mask'') and the input filename (typically ``mask.slits''). You will then be asked to press the ``b'' key at the left and right edges of the slit corresponding to your first object, followed by ``q'', then the ``a'' key at the center of slit, followed by ``q''. This will be repeated for each alignment slit. After the last slit, the calculated positions of the objects will be marked on the screen, and an output file called ``mask.ap'' will be produced which will be used during the night for alignment.

Other Comments

If you wish to sample a wavelength range that is not centered on the standard wavelength of the grism, this can be accomplished by moving all of the slits up or down (in the y-direction) in the slitmask field-of-view. This is typically not a reasonable course of action for the 300 l/mm grism, since its spectral coverage includes nearly the entire optical spectrum. But it may be of interest for users of the 600 l/mm grating (central wavelength is 5000 ┼). For the latter grism, the spectral scale of 1.6 ┼per pixel, the field-of-view in the y-direction of , and the scale of 0.4'' per pixel, result in an available range of ┼for the central wavelength. (With this grism, the full spectral coverage is approximately 3200 ┼.) You will need to avoid the label in the upper-left portion of the mask, and should be advised that any distortions in the COSMIC field-of-view will be more pronounced as the slits are placed further from the center of the field. Moving the slits upwards (in the positive y-direction) will shift the center wavelength towards the red. This can be done in the ``cosmicslitmask'' program (option 10) which you use to create the slitmasks.

The scale of your input object coordinates should be known quite accurately; using a ``word-of-mouth'' CCD pixel scale, or one listed in a handbook, might not be as accurate as you might think. The best solution is to observe a globular cluster with the same CCD setup as you observed your multislit object field, and then use astrometry programs to get a good plate scale. Better yet, you can use many stars (> 100) from the cluster to get a mapping of the CCD distortion, and correct your object coordinate list for this effect. For illustration purposes, a 0.5% error in the plate scale will cause a shift in the slits by 1.2'' for a 4' wide set of multislits, which is comparable to the typical slit width of 1.5''. Accuracy of 0.2% or better in the plate scale is a reasonable goal.

The focusing of the CCD camera on the apertures is dependent on temperature. This adjustment is not typically done by the setup crew, but should be considered by the observer. Measure the slit widths (FWHM) in the afternoon and determine if they appear to be in an acceptable focus. The focus procedure requires two people, one in the cage turning screws and the other at the computer taking images and measuring FWHMs. Contact experienced individuals before attempting this procedure, as mistakes can be destructive to the alignment of the instrument. It is also important to note that the optimal focus for the longslit and for your multislits might be different, so you may have to choose which one to optimize.


next up previous
Next: Procedure for Aligning the Up: Observing with Multislit Masks Previous: Observing with Multislit Masks
Roy Gal
2/11/1999