[daip] [!4715]: aips - SNP2D

[Lynn D Matthews]

Lynn D Matthews updated #4715
-----------------------------

SNP2D
-----

           Ticket ID: 4715
                 URL: https://help.nrao.edu/staff/index.php?/Tickets/Ticket/View/4715
           Full Name: Lynn D Matthews
               Email: lmatthew at haystack.mit.edu
             Creator: User
          Department: AIPS Data Processing
       Staff (Owner): -- Unassigned --
                Type: Issue
              Status: Open
            Priority: Default
                 SLA: NRAO E2E
      Template Group: Default
             Created: 24 March 2014 12:05 AM
             Updated: 24 March 2014 12:05 AM
                 Due: 26 March 2014 12:05 AM (2d 0h 0m)
      Resolution Due: 01 April 2014 12:05 AM (8d 0h 0m)



I am working on a 3-bit "continuum" data set where a few channels contain a bright maser line that I would like to use to calibrate the phases across the entire band.

I have for the first time attempted to use the task 'SNP2D' for this purpose, but have encountered a few issues. 

(1.) When one loads the task, there is a typo in the name (it says 'SNCOP' instead of 'SNP2D').

(2.) The help file notes that the rest frequency must be set first with SETJY. Because doppler setting was not yet enabled for 3-bit mode when I took my data, the initial rest frequencies and velocities in the SU table were all 0. 

To fix this,I assumed that the channel where the maser peaks (channel 93 of IF 10) corresponds to the source velocity of 42.0 km/s LSR and the rest freq. is 43.120 GHz. I then ran SETJY with the following:

OPTYPE ' ' ; SOURCE 'Maser star'
BIF 1 ; EIF 0
NIF 10;  SYSVEL 42.0
RESTFREQ 4.312e10, 0
VELTYP 'LSR' ; VELDEF 'RADIO'
APARM(1) = 93

The SU table then showed:

COL. NO.              18                     19
     ROW      LSRVEL                 RESTFREQ
  NUMBER      M/SEC                  HZ
       1      -7.3276659561D+06      4.3120001024D+10
       1      -6.4377440293D+06      4.3120001024D+10
       1      -5.5478221025D+06      4.3120001024D+10
       1      -4.6579001758D+06      4.3120001024D+10
       1      -3.7679782490D+06      4.3120001024D+10
       1      -2.8780563222D+06      4.3120001024D+10
       1      -1.9881343954D+06      4.3120001024D+10
       1      -1.0982124687D+06      4.3120001024D+10
       1      -2.0829054190D+05      4.3120001024D+10
       1       6.8163138487D+05      4.3120001024D+10
       ...
etc., etc.

It seems that the resulting velocities in the SU table are "backwards"---i.e., since frequency increases with IF number, they should go instead from larger (positive) to smaller (negative).

(3.) Ignoring the above, I created two SPLIT files: one containing my target source and the other containing a single channel for self-cal purposes. I ran self-cal, creating an SN table on the single channel file. 

Next, I ran SNP2D with the single-channel file as input, the multi-channel file as output. 
However, all of the delays in the newly created SN table are "NaN". 

If I instead use the single-channel file as both input and output, then sensible delays are computed (but with entries only for a single IF).

Finally, I tried to run self-cal by specifying a single channel on the multi-IF file. However, this turned out to be completely impractical because of the large data size. I was still waiting for CALIB after more than an hour and had to give up.

Lynn

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