[daip] [!7226]: AIPS - strange scaling of visibility amps & weights w/ DOBAND=1

[Michael Bietenholz]

Michael Bietenholz updated #7226
--------------------------------

              Status: Open (was: Closed)

strange scaling of visibility amps & weights w/ DOBAND=1
--------------------------------------------------------

           Ticket ID: 7226
                 URL: https://help.nrao.edu/staff/index.php?/Tickets/Ticket/View/7226
           Full Name: Michael Bietenholz
               Email: mbieten at yorku.ca
             Creator: User
          Department: AIPS Data Reduction
       Staff (Owner): Eric Greisen
                Type: Issue
              Status: Open
            Priority: Default
      Template Group: Default
             Created: 08 September 2015 05:34 PM
             Updated: 07 October 2015 02:10 AM



Although it does in fact behave as you suggest, I still think its a bit odd that you can apply a normal-seeming bandpass correction and have the weights of the visibilities for some antennas change by an order of magnitude compared to the others.

The reason for the counter-intuitive (to me anyway) behaviour is the scaling of the weights by the mean BP gain squared.  With one or two channels with BP amplitudes down near 0 (which is not that unusual) the *mean* gain can become arbitrarily large, even though most of the channels have BP-gains (and thuse weight-scale-factors) near unity, and so these same near-zero channels can drag the weight for the whole IF down substantially.  

I still think the weights should probably become channel-dependent at the point of BP calibration - is AIPS limited to requiring the weight to be the same for all the channels in an IF?  With channel-dependent weights the above situation doesn't happen, and the overall weight of an IF doesn't change very much upon BP assuming that most of the channels have weights near unity.

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