[daip] [!8537]: AIPS - VBGLU

[Eric Greisen]

Eric Greisen updated #8537
--------------------------

              Status: Closed (was: Open)

VBGLU
-----

           Ticket ID: 8537
                 URL: https://help.nrao.edu/staff/index.php?/Tickets/Ticket/View/8537
           Full Name: Lynn Matthews
               Email: lmatthew at haystack.mit.edu
             Creator: User
          Department: AIPS Data Reduction
       Staff (Owner): Eric Greisen
                Type: Issue
              Status: Closed
            Priority: Default
      Template Group: Default
             Created: 09 June 2016 06:35 PM
             Updated: 10 June 2016 04:31 PM



I now understand more and some of the problem lies in your
input FITS IDI files.  The ANTENNA table says that there are no
polarizations with calibration values and yet gives values as if
there were 1 - i.e. 31 values.  So FITLD builds a table saying there
are no polarizations but having 31 values in the AT table and
62 values in the AN table, but still NOPCAL=0.

VBGLU looks at NOPCAL and sees that there are none and so
builds 62 correct (value 0) plus 62 uninitialized values.  If
you use PUTT verb with keyword 'NOPCAL' and keyval=2,0
on the two input files to VBGLU then it works correctly.

I have link edited all the many tasks affected by the buffer
size error.  Tomorrow's MNJ should fix the problems with the
output file from VBGLU.  

I do not know what version of DiFX produced these IDI files.
If you can, reporting this anomaly to those involved would be good.


Eric Greisen




------------------------------------------------------
Staff CP:  https://help.nrao.edu/staff