[evlatests] Good News, Bad News ...

[Rick Perley]

    I reported yesterday on an array of failures which afflicted a test 
from the previous night. 
    I re-ran that file last night, but put in a 'wrap' card to prevent 
the long azimuth slew. 

    First, the Good News:

    1) There were no missing BDF files, and all requested scans provided 
data. 
    2) All requests for referenced pointing were honored, and solutions 
for most antennas were obtained for each requested observation.  I write 
'most' antennas, because ...

    But not all is good, as:

    The failures modes described yesterday -- wherein the majority of 
the scans have ~50% of the data blanked out (pure zeros) -- afflicted 
the test last night as well.  The two modes (essentially, antennas are 
blanked out in pairs, with mode 'A' being antenna 1 and 2, 5 and 6, 9 
and 10, etc., and mode 'B' being the opposite (antennas 3 and 4, 7 and 
8, etc.) are still present, and afflicting nearly all the scans and all 
the bands. 

    Something is seriously amiss with the correlator.  Do we know that 
this problem is not affecting the science data? 
    My setup is completely vanillla -- eight subbands of 128 MHz each 
(except at L-band where it is 64 MHz) for each of the two IF pairs.  
This configuration has been run successfully an almost countless number 
of times.