smdfile.txt
smdfile.txt — Plain Text, 6 KB (6259 bytes)
File contents
The Slit Mask Definition File The "maskgen" program produces a Slit Mask Definition File, which has a name of the form "<maskname>.SMF". This file is designed to contain all the information necessary to construct a mask, and to observe with that mask. It is used as input to the mask cutting program, and to utilities which display or plot an image of the mask. It is also designed to be available to data reduction to allow processing programs to predict the spectral data on the detector from the slit positions at the mask location. Data in this file are in the following groups: Identification: The mask name, and observer name are used to identify this mask; they are engraved on the rim of the mask itself when it is cut. A commentary title is used by the observer to describe exactly what this mask was made to observe. The date of creation, version of the mask generation software and a unique mask identifier are also included in a comment for the LCO slitmask database software. This section also states which instrument the mask is made for, and that will show which kind of physical mask is being described. Celestial positions: The position to which the telescope must be pointed, and the rotational angle of the I.M.A.C.S. instrument so that the mask features will correspond to their intended celestial objects. Also, the celestial coordinates for each object are repeated. A celestial position is not present for a position-independent mask, such as a standard long slit, or a pinhole test mask. Useful comments: A comment on rotator flip, i.e. whether the rotator passes through +/- 180 degrees while the object is above the horizon is produced. A comment containing part of the observing catalog entry for the observation is given, prefixed by "!.OC". This will enable a script to read all the .SMF files for a run, and produce an observing catalog file for use at the telescope. For the observing catalog particulars, see: http://www.lco.cl/telescopes-information/magellan/instruments/observing-catalogs The information following the !.OC introducer is the fields 2-15 of the catalog, prefix an entry number/name to these to get a line for your observing catalog. This will produce the correct instrument rotator mode and offset value for the mask position. The "obscat" utility uses these comments to automatically generate an observing catalog for you. Instrumental configuration: The configuration of I.M.A.C.S. for the observation, which disperser is used, grating angle if used, which internal focus is used, the assumed wavelength range and any other set-up values for the instrument. Mask features description: For each feature cut into the mask, its exact position, shape and size is given to enable the cutting program to cut that feature, and for data reduction programs to determine exactly where the holes in the mask are. There are 3 possible object keywords -- SLIT, HOLE, AREA. For each there are several fields separated by whitespace. Those keywords are written as "Slit", "Hole" and "Area" for virtual objects which are present logically, but not actually cut into the mask itself. After the keyword, there is an object name. The position of the object, if present, is in hh:mm:ss.ss -dd:mm:ss.ss format. If the object is position independent, the position may be omitted, or more usually given as "--- ---". Following the optional position field is the mask location of the object in mm (X and Y). Then there is a width in mm as a decimal number. For the HOLE type, and optionally for the SLIT type there is an integer shape code (0=circle, 1=square, 2=rectangle, 3=special), then the length in -x and length in +x, followed by a tilt in degrees of counterclockwise rotation (on the sky). When applying these dimensions to an IMACS mask, the x coordinate and rotation are reflected because the coordinate system of the cutting machine applies to the "back" of the mask, having the opposite sense of the sky view. Both SLIT and AREA are of the rectangle shape, so they have no shape code in their data. The AREA object will be cut differently, with a better algorithm for edge of mask truncation and with gaps in the cutting so the inside piece does not fall out while being cut. In summary then, the object data consists of: keyword object [RA DEC] x y width * length- length+ angle * = shape code (integer) may be optional. Type HOLE always has integer shape code, SLIT may have integer shape code, AREA has no shape code and is always a rectangle. The shape code is distinguished by having no decimal point, while the width and lengths are always written with a decimal point. There are some special processing options for keywords having an exact form before being rendered case-independent: "Slit" = Virtual object; marked as special in data, not cut in mask "Hole" = Virtual object; not cut in mask "AREA" = R.A. and Dec. present in line, otherwise omitted New special processing options may be defined in the future, and new object keywords may be defined too. Existing data conforming to these format rules should always read in correctly. Objects of type SLIT are regular objects, HOLE is always a reference object, and AREA is the special area object. Some special shapes have been defined for test masks, and these are not available for observing masks. Additional shapes or extensions to the existing data may be defined in the future if needed. It is important to note that the .SMF file is automatically written by the "maskgen" task, and read by several programs including "maskcut" and the data reduction programs. The latest version of the reading program will always be made to read any file produced by the latest version of the writing program. Files which have not been written by "maskgen" or which have been altered will be rejected by the "maskcut" task; that is why the checksum record appears at the end of the file. Since the "maskcut" program insists on processing only data from the "maskgen" program, those special programs used to produce test masks or special masks will usually write a MADE line specifying MaskGen as the generating program. When this is done, comments giving the true origin of the file will be provided.