[evlatests] More Referenced Pointing Results

[Rick Perley]

    As mentioned by Ken, another referenced pointing test was done this 
morning. 

    The differences between this test and last evening's test are:

    1) The referenced pointing was done at X-band,
    2) The file was in DURATIONS. 

    The sequence of operations was the same:
   
    K band, 'dummy', no ref applied
    K band, no ref applied.
    X band reference solutions
    K band, ref applied
    K band, no ref applied
    X band reference solution
    Q band, ref applied
    Q band, no ref applied
    X band, reference solution
    K band, ref applied
    K band, no ref applied.

    Results:

    1) None of the scans which preceded a reference pointing scan 
provided any data!  We had thought this known bug had been fixed -- 
clearly it has not!  Note -- this problem only applied to files in 
durations. 

    2) As the final pair of scans were observed (there being no pointing 
scan following the final one), we can test for the application of the 
referenced pointing.  *** There is no evidence that any referenced 
pointing solution was applied to any antenna! ***  As the reference 
pointing solutions are large (typically 0.5 arcminutes, not surprising 
given the sunny skies and low sun angle), the amplitude difference would 
have been obvious.  There is no change of amplitude with, or without, 
referenced pointing. 
    Note that this behavior is quite different than yesterday's -- where 
I claim that for stoptime files, the solutions ARE applied to EVLA 
antennas, and not (at least in some cases) to VLA antennas. 

    3) More bad news:  There was a global phase jump between the last 
pair of scans -- referenced to an EVLA antenna, all VLA antennas 
'jumped' by about 80 degrees.  Both polarizations of both IFs jumped by 
about the same amount.  All EVLA antennas remained at their prior 
levels.  There was no change of band, frequency, of Fluke setting, 
between these scans. 

    4) Antenna 18, IFs A and C, have their phases jumping up and down, 
by about 80 degrees.  The duration of the 'jumped' state is 10 seconds, 
exactly. 


    <Audible Sigh>  We're not out of the woods, yet. 

    We will try run this file again, but with stoptimes, when time is 
available.