[daip] [!8252]: AIPS - RLDLY: output SN-table phases not consistent with delays (DOIFS=1)

[Eric Greisen]

Eric Greisen updated #8252
--------------------------

RLDLY: output SN-table phases not consistent with delays (DOIFS=1)
------------------------------------------------------------------

           Ticket ID: 8252
                 URL: https://help.nrao.edu/staff/index.php?/Tickets/Ticket/View/8252
           Full Name: Michael Bietenholz
               Email: mbieten at yorku.ca
             Creator: User
          Department: AIPS Data Reduction
       Staff (Owner): -- Unassigned --
                Type: Issue
              Status: Open
            Priority: Default
                 SLA: NRAO E2E
      Template Group: Default
             Created: 05 April 2016 12:59 PM
             Updated: 05 April 2016 09:28 PM
      Resolution Due: 13 April 2016 12:59 PM (7d 15h 30m)



I have now run RLDLY on an EVLA data set I had.  After FRING (and I ran a recent version)
the RR and LL phases line up reasonably but the RL and LR phases have jumps at every 
IF boundary.  After RLDLY the phases line up quite well across the IF boundaries incl
the one between IFs 8 and 9 where different delay solutions are required.  I agree
that the listed phases do not seem to bear a relationship to the delays being added
(my case delays of 9 and 21 ns, phase diffs of -79 and +109 in the 2 halves).

If I change the code to simply compute phases as (F(if)-F(1)) times delay, then my data set will
not line up well at all in the RL and LR sides and CALIB cannot change that.  I will continue to 
explore because I agree this is odd.

Eric




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