[daip] [!9168]: AIPS - clcal introducing amplitude / gain errors

[NRAO Helpdesk]

Amy Mioduszewski updated #9168
------------------------------

       Staff (Owner): Amy Mioduszewski (was: -- Unassigned --)

clcal introducing amplitude / gain errors
-----------------------------------------

           Ticket ID: 9168
                 URL: https://help.nrao.edu/staff/index.php?/Tickets/Ticket/View/9168
           Full Name: Alexandra Truebenbach
               Email: alexandra.truebenbach at colorado.edu
             Creator: User
          Department: AIPS Data Reduction
       Staff (Owner): Amy Mioduszewski
                Type: Issue
              Status: Open
            Priority: Default
                 SLA: NRAO E2E
      Template Group: Default
             Created: 18 October 2016 06:18 PM
             Updated: 18 October 2016 06:18 PM
                 Due: 20 October 2016 07:59 PM (2d 0h 0m)
      Resolution Due: 26 October 2016 06:18 PM (7d 22h 19m)


Hi,

I'm reducing a VLBA data set, using the same steps as http://www.aoc.nrao.edu/~amiodusz/sumschool16/vlba_tutorial.shtml, the VLBA tutorial from the summer school this year.  Everything goes fine until I get to clcal to phase reference my target.  The resulting cl table has large outliers in gain (see snplt.png below), which translates to large, unrealistic amplitudes seen in editr (afterclcal.png) and in the uv plot (uvplt.png).  I've tried the same process with another data set and I'm getting similar issues.  Is this how the gains are supposed to look after clcal?  I'd think no given how they jump around don't vary smoothly with time.

Some other info:
I've looked at all previous cl and sn tables with snplt.  They all look fine.  Most have gains ~1 except for the sn table after running VLBAAMP, which has gains smoothly varying ~ 20.  For clcal, I'm running the default with the following other parameters: Sources ‘target’, calsour ‘phase reference’, interpol ‘simp’, snver 0, invers 0, gainver 0, gainuse 0, refant 9.
I've also tried changing the interpol to '2PT' in case it was some error with simple interpolation.

Thanks for your help,
Alex Truebenbach


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