Date: Thu, 6 Mar 2003 11:49:40 -0500 (EST)
From: David Schlegel
Subject: Bad fibers, bad CCD regions, etc
Scot et al,
What Scot is pointing too is indeed bad regions of the CCD.
In particular, the r2 CCD has groups of bad columns.
This is what caused the famous "fiber 375" problems years back.
Plotting the reduced spectra shows the "bad pixel" mask set
in these cases.  You can see the bad pixel regions march
through different wavelength regions in neighboring fibers
(since the spectra are curved on the CCD).
Note that the tools to look at bad fibers / bad CCD regions
all exist.  There's certainly no need for more code.  You
can grep through my postings to sdss-obs, sdss-spectro and
the bug database ("sop" and "procedures" categories, many
of which have been closed).  For some reason, none of this
ever seems to make it into procedures documents.  I find
this very disturbing, as should you.
But, for the question at hand, look at:
  PLUGHISTORY - Plot the history of unplugged fibers
  ATVRAWSPEC - Display a 2D spectroscopic image with bad columns in red
Note that all these things can be run on either sdsshost.apo
or sos.apo.
-David