Don York has posted to these mailing lists about apparent flatfielding problems with the echelle. These problems manifest themselves as a medium scale fringe-like pattern in extracted spectral orders, which is easily seen in tungsten-quartz lamp exposures and is less apparent in stellar exposures. The ripple is characterized as an undulation of amplitude 1-2% about the true echelle blaze function on spatial stretches of 10 to 20 pixels, thus limiting the maximum signal-to-noise ratio on this length scale. Possible optical and mechanical explanations for the ripple were proposed, tested, and eliminated as contributing factors. The ripple has now been demonstrated conclusively to be an artifact of the default order extraction method used in IRAF's echelle software package. In particular, the algorithm used in aptrace does not cope well with minimal sampling of spectral orders in the direction perpendicular to the dispersion. Ripple is introduced in regions of strong order slope or curvature due to aliasing. While ultimately a detector with smaller pixels would reduce the observed effect, a software fix to the relevant IRAF packages is definitely warranted but may require some time to implement. Don and I will keep everyone posted. Julie Thorburn APO APO APO APO APO Apache Point Observatory 3.5m APO APO APO APO APO This is message 13 in the apo35-echelle archive. You can find APO the archive on http://www.astro.princeton.edu/APO/apo35-echelle/INDEX.html APO To join/leave the list, send mail to apo35-request@astro.princeton.edu APO To post a message, mail it to apo35-echelle@astro.princeton.edu APO APO APO APO APO APO APO APO APO APO APO APO APO APO APO APO APO