[daip] [!10670]: AIPS - NORMALIZ in CALIB

[Lynn Matthews]

Lynn Matthews updated #10670
----------------------------

NORMALIZ in CALIB
-----------------

           Ticket ID: 10670
                 URL: https://help.nrao.edu/staff/index.php?/Tickets/Ticket/View/10670
           Full Name: Lynn Matthews
               Email: lmatthew at haystack.mit.edu
             Creator: User
          Department: AIPS Data Reduction
       Staff (Owner): -- Unassigned --
                Type: Issue
              Status: Open
            Priority: Default
                 SLA: NRAO E2E
      Template Group: Default
             Created: 12 July 2017 06:37 PM
             Updated: 12 July 2017 06:37 PM
                 Due: 14 July 2017 06:37 PM (2d 0h 0m)
      Resolution Due: 20 July 2017 06:37 PM (8d 0h 0m)



Some questions/comments on the NORMALIZ adverb in CALIB.

I have long thought that setting NORMALIZ=1 during an 'A&P' self-cal was the equivalent to what was formerly accomplished using CPARM(2)=1 and that this is desirable to keep the amplitudes from drifting.

However, I have a data set where I noted a significant drop in the flux density (30%) when I use NORMALIZ=1, CPARM(2)=0, so I did some further tests.

First, contrary to my reading of  the HELP file, NORMALIZ=0, CPARM(2)=1 is not the same as the former CPARM(2)=1. In fact, this does not appear to do any normalization. No messages about normalization are printed, and  MGMOD    =  1.0000000000000D+00 in the SN table header.

Second, I find NORMALIZ=1, CPARM(2)=1 does not result in the same loss of flux.  If I understand correctly, this is enforcing a median rather than a mean of the gain modulus. But is this the right thing to do? I could not find any real discussion of this in the help file or Cookbook.

Thanks,
Lynn


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