[evlatests] Reference pointing tests

[Rick Perley]

    Summary:

    I'm pleased to report that EVLA reference pointing seems to work 
perfectly. 

    Some details:

    I ran tests last night, under excellent weather conditions, which 
tested both
'primary' reference pointing, and 'secondary' reference pointing. 

    The major anticipated results were seen:

    a) All 3 antennas gave good solutions at X-band with primary reference
pointing.  The errors were too small to note any resulting improvement in
gain.

    b) Antenna 13 gave no solutions at K or Q bands.  It would have been
amazing if it did, as it has no receivers at these bands!

     c) Antennas 14 and 16 obtained solutions at K-band, both in 'primary'
and 'secondary' referencing.  I have not inspected the solutions yet, but
for both antennas, there is a clear improvement in gain and sensitivity
when the solutions are applied.  With the application, both antennas are
amongst the best in sensitivity, as judged by 'aipsweights'.  (Single IF
exceptions are seen, and will be noted in a subsequent note).

    d) Only antenna 14 gave solutions at Q-band.  Antenna 16 is
fringing, but very weakly.  It seems likely that its collimation is so large
that the raster failed to find the beam.  For antenna 14, a clear 
improvement
in gain and sensitivity is seen with the reference solutions applied.  
Antenna
14's sensitivity is only fair, however -- not the worst, but probably 
not as good
as the median VLA antenna.   I speculate this is an efficiency or Tsys 
issue,
not a pointing issue.  (As expected, antenna 16's sensitivity is very bad).