[evlatests] Some observations in OSRO

[Rick Perley]

    Ten hours on Cyg A, plus calibrators, were taken Friday night in 
OSRO 1 mode.  I switched around four bands:  L, S, C, and X.  No 
referenced pointing.   I changed band with every new observation (so I 
did not try source-cal-source at one band). 

    A brief review of the data shows a few items of interest:

    1) about 4% of the data in the X-hands are 'integer zeros'.  Nearly 
all of these are associated with the beginning of scans, typically a few 
seconds long, but on occasion, as long as a minute!  There is never any 
obvious pattern of which antennas took a minute to show up.  Rarely, a 
short (few seconds) block of data, seemingly antenna-based, 
'disappeared' (were identically zero). 

    2) Antenna 25, IF 'D' had essentially zero amplitude fringes at all 
bands.  Yet, the bandpass looked fine!  Perhaps the power to the 
samplers was far too low?  Michael R. thinks this problem has been 
repaired ... has it? 

    3) All data were taken with 'set and remember' at all IFs.  Yet the 
variation in the fringe amplitude of the calibrator source 2007+404 at 
L-band is as high as factors of 2 or more, from scan to scan, on some 
antennas.  This does not seem to be an issue at the higher frequency 
bands.  Nor is this a problem for 3C286 at L-band.  I believe this is 
due to the influence of Cyg A, which is well over 20 dB stronger than 
the calibrator, and only a degree or so distant -- the spectrum shows 
strong beats in frequency, so the formation of a 'channel 0' for 
calibration purposes will reflect these if the averaging width (in 
frequency) is not an integer multiple of the beat -- which it never is, 
in general.  I'll be able to check this suggestion when I'm free of the 
summer school ...  (The mystery to me is why I don't recall this being a 
problem with the VLA -- we used 50 MHz BW, which would certainly have 
been subject to the same issue). 

    Some antennas (notably 8 and 18) are showing a complete lack of 
'returning to phase' at L-band.  (I haven't been able to check this at 
the other bands yet).   Each observation of the calibrator, for these 
antennas, has a completely different phase.