[evlatests] re AT343, K band.

[Vivek Dhawan]

A bit more on this after closer inspection:

Flagging: EA 14 was completely flagged, and deserved to be -
           no fringes. I do not know the reason (FRM??).

           EA22 was also flagged out, but did have fringes.

           I did not notice these earlier because the whole
           antenna was missing!

           EA13 actually had 3 chunks of data flagged of about
           30-40min each, which did deserve to be flagged.

Source names: The observe file had the following structure:
           one scan on a calibrator, with calcode, followed by
           a block of fast switching using the same calibrator.
           Same general thing on 4 or 5 blocks, different targets.

           In the final dataset, each one-of scan had a different
           source name from the same cal in the fast-switch loop.
           The coordinates were identical, but the name differed
           by 0.1 degree in dec. The calcode did not get attached
           to the latter.  After a quick talk with Eric, he told
           me about AIPS task DSORC which will unify siamese twin
           sources. I do not know if this will cause problems
           downstream.


V.




Earlier message:

K band continuum. 3.3s integrations.

Flags are working correctly, the scans on calibrators are often
only 27s long total, and most of the times/baselines 13s of good
data survive on these scans. From closer looks at some stretches
of data, it appears that very little good data is being flagged.

Exceptions were EA13, which had 2 chunks of about 30min flagged;
and VA2, which had only 8-10s of data on most of the short cali-
brator scans. In spite of this I think the data will be quite
usable, the cals are strong enough. However, the weather was
not great.

Phases were wobbly, but nothing that looked like the 1m
difference between E/VLA was manifest.

If they are doing the same thing again tomorrow, it should work.
It wouldnt hurt to lengthen the calibrator scans by 10s. I'm not
volunteering to hack their file.

Reference pointing: Judging from the amplitudes, things were
good. Details of the solutions I think Rick was going to look
at.

V.