[daip] CALIB bug in 31DEC04 at the AOC

[Stockdale, Christopher]

Dear dAIP,
  I have run into a very odd CALIB bug in TST (31DEC04) at the AOC (zaurak).
(user #2284)
 
> default calib
> indi 1 ; getn 26
AIPS 2: Got(1)   disk= 1  user=2284   type=UV   20040422.X BAND.1
> flagver -1; docal -1
> calsour '1331+305',''
> go calib
CALIB2: Task CALIB  (release of 31DEC04) begins
CALIB2: CALIB USING 20040422     . X BAND .   1 DISK=  1 USID=2284
CALIB2: Selecting the data
CALIB2: Doing cal transfer mode with point model for each source
CALIB2: This is not self-calibration
CALIB2: Dividing data by source flux densities
CALIB2: Determining solutions
CALIB2: Writing SN table    1
CALIB2: Found           54 good solutions
CALIB2: Failed on       50 solutions
CALIB2: Adjusting solutions to a common reference antenna
CALIB2: Appears to have ended successfully
CALIB2: zaurak       31DEC04 TST: Cpu=       0.1  Real=       0
AIPS 2: Resumes

LISTR/GAIN shows that this has worked fine for RCP in both IFs (this is a standard VLA
continuum experiment at X band), while there are no LCP solutions whatsoever.

> version 'new'
> go calib
AIPS 2: Found in Version=NEW
CALIB2: Task CALIB  (release of 31DEC03) begins
CALIB2: CALIB USING 20040422     . X BAND .   1 DISK=  1 USID=2284
CALIB2: Selecting the data
CALIB2: Doing cal transfer mode with point model for each source
CALIB2: This is not self-calibration
CALIB2: Dividing data by source flux densities
CALIB2: Determining solutions
CALIB2: Writing SN table    2
CALIB2: Found          104 good solutions
CALIB2: Adjusting solutions to a common reference antenna
CALIB2: Appears to have ended successfully
CALIB2: zaurak       31DEC03 NEW: Cpu=       0.0  Real=       0
AIPS 2: Resumes

and now we have everything!

Surprisingly, this is only a problem for 1331+305 -- the other sources work equally
well (in 31DEC04) in both polarizations.  1331+305 is the 4th (and last) source in the
SU table.

If you'd like to look at this data set, you can snag it at
   /home/zaurak/AIPS/DA01/FITS/calibdata.fits

Thanks very much for looking into this --

                     Chris & Michael