[evlatests] EVLA-specific phase jumps

[Rick Perley]

    The winds last Tuesday prohibited any attempt to do precision optics 
alignments, so I used 6 hours of software time to continue exploring 
polarization behavior at L and C bands.  Results (which are outstanding) 
will be presented elsewhere.  The observations were of 5 minute duration 
for each band. 

    Here I report on some peculiar EVLA phase jumps on antennas 25 and 
26, only. 

    The observations were made in spectral line mode PA at a single 
bandwidth of 6.25 MHz.  In general, phase behavior for all antennas was 
excellent.  (Indeed, because the Fluke settings were the same for both 
bands, there were no EVLA-VLA global phase jumps, either). 

    However, two EVLA antennas showed a number of phase changes, and 
phase jumps.  For both, the jumps were common to IFs A and C.  (Because 
I was using mode PA, there are no data on the B/D IFs). 

    Antenna 25 showed jumps at both bands: 
    At C-band, there were 5 easily visible changes.  Three of them were 
accompanied by a 10% rise in correlation coefficient.  For these, the 
timescale of the change was 10 seconds.  The other two jumps were 
instantaneous, and there was no change in correlation amplitude.  All 
these jumps occurred in the middle of various 5-minute observations. 
    At L-band, there were four jumps -- one which occurred in the middle 
of a scan, and three which affected entire scans.  For the situation 
where the phase jumped (by 100 degrees) in the middle of the scan, the 
change took exactly 10 seconds, and there was no change in amplitude. 

    Antenna 26 showed 5 jumps at L-band only.  One of these affected a 
whole 5-minute scan, the others occurred in the middle of the scan.  For 
these, the change in phase took exactly 10 seconds to execute, and there 
was no change in amplitude.  Four of these phase changes were by 140 
degrees, the other, 100 degrees.