DRS

Problems Tracing Orders

The following is two reductions of the same data set (2×ThArNe, 5×FF & 1×60sec exposure of HR1544) but reduced with two different initial guesses. The first set are reduced with the ThAr9857 guess, while the second are reduced with the ThAr9001 guess. The bad data reduction and terrible extracted spectra result from the poor order location and tracing. The ThAr9857 guess was the "standard" guess from the ESO-1.52m, while the guess ThAr9001 was created specifically to overcome this problem which has occured from time to time since FEROS was moved to the MPG/ESO-2.20m telescope. The problem is caused by the intrinsic change in cross-dispersion profile from day to day and by a one off apparently systematic change which probably occured when FEROS was moved.
The bad tracing is subtle, but clearly visible in the display at the normal scale during the DRS initialisation. The Zoomed in display is created with the following MIDAS commands:

MIDAS> load/imag FF<FFNum> sc=-3
MIDAS> load/tab {CENTER_TBL} :Fit_pos :Ypos ? 0 1

The plots show most of the full range (4000-9000Å), H[alpha] and H[beta].