[evlatests] Quick Test Results -- not good

[Rick Perley]

    Following an X-band pointing correction made by Vivek and Barry this
afternoon, I ran a quick check of all EVLA antennas at L, C, X, and K bands.

    For each, I did a pointing scan on 3C84, followed by a short on-source
integration to check sensitivity, etc. 

    Results are not encouraging:

    1) Pointing.  No pointing solutions were obtained at any band for 
antennas
14 and 16.  Antenna 13 gave only one solution (a very odd one ...) at 
L-band,
and none at any other band.    Antenna 18, which purports to work at 
X-band,
gave no solutions. 

    2)  Amplitudes.  Although stable fringes were seen on all IFs for 
antennas
13, 14, and 16, all amplitudes are very low.  For antenna 13, they are a 
factor
of 3 to 10 too low.  For antenna 14, the factor is 20 at L-band (all 
IFs) and
3 to 5 at all other bands.  For antenna 16, it's a factor 15 to 20 for 
all bands,
all IFs. Antenna 18 gave no data at all at X-band. 
       However -- the situation is not as simple as above.  The EVLA-EVLA
baselines do not follow the rules above -- they have amplitudes usually much
higher than the geometric mean of the attenuation factors listed.  
Hence, they
have what we call in the business 'large closure errors'. 

    The low amplitudes might be suspected to be the cause of the failed
solutions -- but this is not sufficient.  I used 3C84, about 15 Jy at 
C-band.
Even with the attenuation of a factor of 3, the amplitude should easily
pass the check in PEEK. 

    The EVLA antennas worked well at C-band earlier in the afternoon, during
the pointing run.  Something bad seems to have happened since then.