[daip] 3C286 U band model file (CALDIR et al.)

[Michael Rupen]

Hi Daip,

  CALIB does not seem to use the model correctly when reading it from
the images available through CALDIR/CALRD.  I believe this is due to
the size of the image to which that model is attached, for reasons that will 
become clear below.

  So here's the problem.  I have an observation of 1331+305 in A config.
at U band.  I run SETJY, then read in the appropriate model with the spiffy
new CALRD, and run CALIB (in OLD or TST, doesn't matter) specifying the CC
table attached to that model file.  Note that the size of that image is
91 x 105 pixels. CALIB seems to correctly scale the CC table, but finds no
solutions.  A point source model on the other hand returns 104 good
solutions.

  Now I read in the U band FITS file from Claire's Web site.  The CC table
is the same, but image is now 512 x 512.  CALIB using Claire's model
attached to Claire's image works fine: 104 good solutions.  So, I copy 
the CC table from the CALRD image over to Claire's image.  CALIB using
the CALRD model attached to Claire's image also works fine: 104 good
solutions.

  So I conclude that the model (CC table) from CALRD is fine, but CALIB
can't handle such a model attached to the CALRD image.  Since that image is
OK apart from being a peculiar size and shape, I assume it's the peculiar
size and shape that are causing the problem.

  The data set involved in all this is Brian Cameron's (Dale's student),
and unfortunately I don't have access to that right now (hence the lack of
inputs/outputs!).  He is currently sitting in Dale's office -- please contact
him if you'd like to play with that specific data set, or I can give you a
similar one.

  Cheers,

            Michael

p.s. The specific inputs to CALIB didn't matter much, so I just did
  DEFAULT CALIB, then set the INNAME/IN2NAME, CALSOUR, NMAP, INVER, and NCOMP.