[daip] [!7729]: AIPS - Large phsae errors with an antenna having different total bandwidth

[Jae-Young Kim]

Jae-Young Kim updated #7729
---------------------------

Large phsae errors with an antenna having different total bandwidth
-------------------------------------------------------------------

           Ticket ID: 7729
                 URL: https://help.nrao.edu/staff/index.php?/Tickets/Ticket/View/7729
           Full Name: Jae-Young Kim
               Email: jykim at astro.snu.ac.kr
             Creator: User
          Department: AIPS Data Reduction
       Staff (Owner): -- Unassigned --
                Type: Issue
              Status: Open
            Priority: Default
                 SLA: NRAO E2E
      Template Group: Default
             Created: 14 December 2015 05:34 PM
             Updated: 16 December 2015 07:01 PM
      Resolution Due: 22 December 2015 05:34 PM (5d 22h 33m)



Dear Eric Greisen and Amy Mioduszewski

I am sorry for lack of my explanations.

There are some stories behind about why we applied UVFRE.
Since correlator output several files with different frequency structures for the entire observations, the files had to be gridded by UVFRE to have same frequency structure to become a desired large file containing everything.
At least for the part that I am analyzing now, however, we can proceed without UVFRE so maybe this is not an important issue at here...

Also thanks for some comments on the data.
First, to answer the question on the manual phase-cal of PdBI, yes, the manual phase-cal was done in multiple steps because the data did not have such a scan.
So we found solutions for (a) VLBA and GBT first and GBT as refant, then (b) connected GBT and Pico Veleta, and finally calibrated (c) EU stations only. 
I think the strange "S"-shaped phase of PV-PB baseline is actually due to bandpass shape of PdBI because it becomes almost flat once BP table from BPASS is applied. 
As an example, I attached a figure with POSSM plots from phase-only calibrated smaller data set which includes only four European stations and only first 8 IFs.

But applying the bandpass calibration did not make a big difference in terms of the final phase noise in individual baselines and the better shape in POSSM came at the expense of some additional uncertainties in the amplitude of the sources. 
For this reason I preferred just flagging first and last a few channels of that station so far.

As for CALIB, if different number of IFs cannot be the main reason, I will first try to downweight PdBI if I could achieve a better final result, e.g. a map with higher DR.
Or can I perform the manual phsae-cal and the global fringe fitting in a bit different way to examine other potential problems with that station?
I can think of running FRING for all stations except PdBI first and then running one more FRING with updated CL table with ANTENNAS=0 and DOFIT='PdBI'.
But I am not sure because, as long as only phase is concerned, the attached POSSM plot already looks pretty good for me if bandpass calibration is applied.

Best regards,
Jae-Young Kim

------------------------------------------------------
Staff CP:  https://help.nrao.edu/staff
-------------- next part --------------
A non-text attachment was scrubbed...
Name: BPASS.APPLIED.ps
Type: application/postscript
Size: 909562 bytes
Desc: not available
URL: <http://listmgr.nrao.edu/pipermail/daip/attachments/20151216/754090f0/attachment-0001.ps>