[evlatests] Stellar high-frequency data

[Michael Rupen]

> Was your test able to say anything about phase transfer at high
> frequencies?
>

* It is unclear whether ref.ptg. is being applied -- Rick was looking
   into that with recent data and found confusing results.  I'm
   assuming for now that this reflects windy weather rather than
   something fundamentally wrong in the system; Ken might comment
   on his recent pointing results which I think looked OK.

* http://www.aoc.nrao.edu/~mrupen/TestPlots/CptgCKobs_vplot.pdf
   shows the results of a 32 minute run involving C-band pointing
   followed by K and C observations with pointing applied.  Greenish is C;
   reddish is K. Interesting bits at K band are:
   - phase jumps on 4, 8, 10, 11
   - loss of fringes for some antennas/IFs for some scans
     (usually all scans following a refptg but I've seen exceptions in
     other runs)
   - ea20 has the pointing problem Ken described and should simply be flagged.
   - there are few to 5% amp jumps on several antennas/IFs
   From these data it looks like you're basically OK if you calibrate
   amp/phase after every refptg scan.  Note however that occasional
   scan-to-scan phase jumps are seen even just sitting at C band.

* We did a similar C-ptg + Ka/C-obs run which Vivek says looked quite
   similar.  Most of the remaining OSRO projects involve Ka band.

* There are indications that using X-band leads to more phase/amp jumps.

Based on this I believe we can observe high-frequency OSRO so long as
observers (1) point up at C band, (2) calibrate carefully after each
ref.ptg. scan (good practice in any case), (3) keep an eye out for full
and partial loss of coherence for individual antennas/IFs.

Of course the weather is making this mostly moot at the moment but we
may get some observations in Sunday night.

              Michael