[evlatests] Poor sensitivity after shutdown likely due to heaters not turned on

[Michael Rupen]

Hello all -

   I reported this morning on a problem with some data taken just after the
Thanksgiving shutdown.  I've now examined those data more carefully,
and Rick has convinced me this was likely due to the heaters not being
turned on for most EVLA antennas.

See below for details.

Cheers,

             Michael

-------
Details
-------

AK679-E-B-D  standard 50 MHz continuum  L, X bands  fast switching throughout
   - observed twice on 24nov07 (dynamic file), separated by almost the full day.
     The first run showed the problem described below; the second (identical)
     run did not.
   - ***The first run was rendered almost entirely useless for the desired
     science (tracking variability of a faint source).

The problem:
   X-band data showed high Tsys (100-200 K, vs. normal 30-60 K) and very low
   sensitivities (factor 10-20 worse than "normal" antennas) on most EVLA
   antennas and one VLA antenna.
   - The offending EVLA antennas were  1  11  14  18  19  21  23
   - The offending VLA antenna   was  28
   - One EVLA antenna (17) had high Tsys (100-200 K), but normal sensitivity
   - Two VLA antennas (3 15) had high Tsys (~100 K),  but normal sensitivity

   The L-band Tsys were fine for all antennas; I've not yet reduced those
   data.

The explanation:
   This is consistent with water/ice on the feeds, presumably due to some
   of the heaters not being turned on.  One expects little effect at L band,
   and a combination of increased Tsys and absorption at X band, leading to
   a loss of sensitivity worse than the bad Tsys would predict -- which is
   what I observed.  The details of the EVLA antennas' hardware differ enough
   that one might get some EVLA antennas acting up, and others not.

   It's not clear why VLA antenna 28 should share in this problem.

--------------------------------------------------------------------------------