[daip] Jagged phases after running DELZN on geodetic data

[Amy Mioduszewski]

Dear Michael,

Hmmm, the multiband delays (MBD) don't look great.  Generally they 
should be large are lower elevations and smaller at higher with some 
scatter (OV in the SNPLT you sent is about right).  So I am worried 
about the MBD, I would start by re-running FRING with a higher SNR 
cutoff (APARM(7) in FRING) or use MBDLY.  To use MBDLY  you have to 
compute the phases in FRING of CALIB first.

I had another thought is there one solution scan?  You have a lot of 
failed solutions and my guess that is because you are observing at high 
freq so a large fraction of geodetic targets are not good at high freq.  
Also do you see a slope in the POSSM plot before you apply the DELZN 
solutions?

Amy

On 6/22/20 10:24 AM, Michael Rosenthal via Daip wrote:
> Good morning,
>
> We are using AIPS to reduce K-band geodetic data for a spectral line 
> dataset. We proceeded through most of the geodetic data reduction 
> steps specified in the CASA guide for spectral lines and astrometry 
> without issue, but identified a problem when plotting the output of DELZN.
>
> For context, we used the source 'J2038+51' as the cal source for 
> VLBAMPCL with no failed solutions. Solving for multi-band delay with 
> FRING yielded 3464 good and 1064 failed solutions, which we thought 
> acceptable. After running the MBD through SNEDT we had flagged out 
> points such that we were left with all points having absolute values 
> <~ 1-2 ns. A screenshot of the SNPLT TV output after running SNEDT is 
> attached here.
>
> The plots of the MBD solutions from TVPL looked reasonable (I've 
> attached two here as an example). We then ran DELZN, when we 
> re-ran POSSM with gainu=6, with CL #6 being the table generated by 
> running DELZN, the phase vs. channel for sources other than J2038+51 
> are not as flat as the example in the CASA guide. I've attached two 
> versions of the POSSM output for the source J0433+05: one using CL #5, 
> from before we ran SNEDT and DELZN, and one using CL #6.
>
> This problem persists if we apply CL #6 to other sources in the 
> geodetic data, and the rest of the sources in the geodetic data have 
> weaker fringes than J2038+51.
>
> Are the solutions we have from DELZN appropriate to proceed with the 
> rest of our data reduction? If not, is there a way we can resolve this 
> issue?
>
> Kind regards,
> Michael
>
> -- 
> *Michael Rosenthal*
> *
> *
> Research Assistant
> Center for Astro, Particle, and Planetary Physics
> New York University Abu Dhabi
>
> _______________________________________________
> Daip mailing list
> Daip at listmgr.nrao.edu
> https://listmgr.nrao.edu/mailman/listinfo/daip

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listmgr.nrao.edu/pipermail/daip/attachments/20200622/32213363/attachment.html>