[evlatests] Some Curious Results from Today's Test

[Rick Perley]

    Michael and Ken took 34 minutes of 4-sub-band data at X-band, on 3C84. 

    Some interesting features are found:

    1) Delays have changed.  With antenna 3 as reference, I find:

    1   -81nsec
    2   -68
    3   0
    9   5
    18   -1
    19   2
    23    8
    24   6
    25   -8
    28   0

                i.e.:  all antennas show very small delays, except 
antenna 1 and 2, which are large. 

    2) There are three 180 degree phase flips.  As usual, there are 5 
flippers, and 5 resters.  The flippers are:  1, 3, 9, 23 and 28. 
The other five are 2, 18, 19, 24 (ref), and 25.   All IFs flip at the 
same time.    The times of the flips are:
       18 46 07
       18 56 57
       19 07 07

    3) Five antennas also had an amplitude 'hop', all at different, by 
adjacent times.  The antennas, and the time of the abrupt change are:

       1   18 47 41
       2   18 48 31
       18  18 45 51
       24   18 45 01
       25   18 42 51

          All hops are the same:  The amplitudes of baselines to that 
antenna abruptly dropped by 10% at the time shown.  There is no phase 
change.  Note that all the changes occur on the first second after a 
10-second 'tick'.  There is a dramatic bandpass change at the same time 
-- see below.

    4) Bandpass solutions were made at every 5 minute interval.  The 
antennas which show the amplitude 'hop' also show a dramatic change in 
bandpass shape -- both in amplitude and phase.  All antennas show 
*superbly stable* bandpasses *after* all amplitude hops were completed. 

    5) Three correlators gave zero amplitudes throughout the run:

       25 x 19, 25 x 24 in sub-band 1
       25 x 24 in sub-band 2.
       The other sub-bands showed good data on all baselines. 

    I'll report on 'wobbles' a bit later.