[evlatests] EVLA Sensitivity

[Rick Perley]

    As part of the 'NSF Construction Report', I've been doing some 
simple tests to make sure we're not exaggerating our successes. 

    One of the report items is whether we've made our sensitivity 
goals.  I have included in the report values for the SEFD which Emmanuel 
generates from the 'stress tests'.  These are based on calibrated 
autocorrelations -- theoretically, these should be fine, but ...  The 
values he provided me were rather lower than I expected to see at some 
bands, so I ran a short test at noon today, observing 3C48 (known flux 
density) and a noise field, 3 degrees away.  Basic calibration was done, 
and histograms of the noise field were made.  Antennas with clearly 
discrepant system temperatures were excluded (generally only 2 or 3 per 
band). 

    The results were so encouraging (read  ... *good*) that I thought 
I'd share them.  Below is a table with the project required SEFD, 
Emmanuel's SEFD, and these new measurements.  Emmanuel's values come 
from mid-band.  Mine are from two, well separated frequencies.  I 
include both, with the lower frequency first. 

Band    Req.         Emmanuel             Rick
-----------------------------------------------------------------------
L              325           342                  330   383
S              235            246                 277    206
C              245            274                 258    207
X              300           237                 218    218
Ku             385           212                 215    231
K               650           402                 325   373
Ka             750            561                518   765
Q             1220           1093             1200  2630
---------------------------------------------------------------------------
NB  Q-band requirement applies to lower end of band.