iraf.net

Hello,
I see that you are a iraf guru so you can probably help me with the following apall (aptrace) problem.Essentially, I cannot get a *perfect* alignment of the lines along the dispersion axis with the aperture aperture based on the trace information. I see this by specifying very small values of llimit and ulimit (e.g. +/-0.1 pixel). This is equivalent to extracting the spectrum along a single line (or rather, a fraction of it). The spectrum extracted in this manner shows a fluctuating continuum, which is evidence that the spectrum does not follow a path of constant signal.Remarkably (and surprisingly), the extracted spectrum does not improve upon increasing the order of the trace of the fitting function. For example, I use Legendre polynomial and I always get the same extracted spectrum for a fitting function order larger than or equal to 2. Yet, I clearly see the improvement of the fit of traced positions when I increase the order of the polynomial. So it seems that this information is "not transmitted" when it comes to extracting the spectra.Thanks for your help
E. Lellouch[color=red:437b63617a][Ed. Note] Forwarded from PM[/color:437b63617a]

I answered this in private mail. The synopsis is that narrow apertures always have wiggles because of the pixel sampling and this is independent of the trace. You should use wider apertures and possibly weighting if there is concern about including low S/N pixels.Frank Valdes