iraf.net

Hi Jeff,The rebin=yes option takes a linearized spectrum and resamples it.
The dispersion function is the linear one in the image header (hence
the defaults are the same). It is independent of any dispersion function
defined via REFSPEC. Note that it does not make sense to
apply a new dispersion function from an arc to a linearized spectrum
unless the arc was also linearized (with exactly the same sampling) and
then a new dispersion function derived from that arc. The is because
the linearization process destroys the relation between the original
pixels and the wavelength at those pixels determined from an arc.
This requires a little bit of thought and can be confusing. Is my
explanation clear?Your choices are to either start over with the unreduced (undispersion
corrected spectra) or USE AN ARC THAT HAS BEEN LINEARIZED IN THE SAME
WAY as the object and treat them as uncalibrated spectra. To force
the programs to return to the case of uncalibrated spectra you must
delete the DC-FLAG keyword with HEDIT.If you need more information let me know.Cheers,
Frank