[evlatests] L-band, and holography check

[Rick Perley]

    While Ken was off having lunch, I snuck in a quick check of system 
behavior at L-band, and a check whether the holography rastering 
problems of last week are resolved. 

    1) L-Band performance. 

    The 3 outfitted antennas (13, 14, 16) ran smoothly and stably.  
Stability in amplitude and phase is the same as any VLA antenna, while 
observing a strong calibrator. 
    The situation w.r.t sensitivity is the same as in past reports.  The 
noise histograms, after proper calibration and adjustment for the 
'closure factor' shows the EVLA antennas are less sensitive than the 
best VLA antennas by a factor of nearly 50% at 1465 MHz, and about 15% 
at 1385
   
    2) Holography. 

    Two down, one to go.  There were three problems noted last week:

    a) The apparent EVLA beams were 20% narrower than VLA beams.  This 
was due to an incorrect stepsize.  This is now corrected.  Beams look 
normal in width.
    b) Starting point.  The EVLA script started with the antennas 'on 
axis', regardless of the requested first position.  This has now been 
repaired, and antennas appear to be where they are supposed to be, when 
the scan begins. 
    c) Center point.  The center point of each scan was clearly one step 
too early at L-band. 
(In other words, in a 13-point scan through the beam, the center point 
should be at location
number 7.  It was always location number 6, independent of which 
direction the scan went).
This problem has not been fixed.  Ken wonders if the actual commanded 
starting point is one step away from where it should be ...

    I'd like to try a couple of variants on the rasters, to see if we 
can change the observed behavior.  all tests so far are 'up-down'.  We 
could try 'left-right', or diagonals.  It would be
good to try 6cm, to see if the offset is stepsize units, or some other 
angle. 

    Looking at the data within each line, some very strange phases are 
seen as we scan
across the source.  But I'd like to get the scans into proper alignment 
before we try making
interpretations ...