setup takes the names of two ucm files which represent two similar exposures but with some sort of change such as collimator position, tilt, or rotation. It also takes a file defining windows. It will then crop each of the two frames to match the windows file, collapse them, measure the positions of peaks and report the change. It gives the positions of the features (assumed in the middle of the window in the collapsed direction).
Only works for single CCD frames, but easily modifiable if necessary. Usage of this routine is as follows: take a suitable setup frame (e.g. an arc which is best for defining focus, rotation and tilt). Using setwin set windows around the arc lines you want to measure. If defining tilt, you would want them as close to the corners of the chip as you can manage; for rotation you might want a couple at each end of a single line in the middle of the dispersed spectrum. This will be the file 'windows' that you need below. Take two exposures with something changing (hartman, rotation or tilt of the chip). Run this routine. Your aim should be to get the position shifts to be as small as possible.
The script generates two files zzz_log1 and zzz_log2 which record the positions measured.
| frame1 | --- | input frame 1 (with or without .ucm) |
| frame2 | --- | input frame 2 (with or without .ucm). Enter 'null' if you don't want a second frame and just want absolute positions. |
| windows | --- | windows isolating the features to be measured (with or without .win) |
| dirn | --- | Direction to collapse frames, X or Y |
| medfilt | --- | Half-width in pixels of median filter, 0 for none |
| fwhm | --- | FWHM to measure peaks, unbinned pixels |
| height | --- | Height threshold for peaks |
This command is a member of the classes: Observing, Scripts, Spectra.
Author: T.R.Marsh
Created: 19 November 2006