[evlatests] Phases and Amplitudes

[Rick Perley]

    Ken and I ran a short fast-continuum file, with a number of
sources, using Barry's modified executor. 

    All phases look good.  No jumps.  Phase behavior is consistent
with normal atmospheric variability. 

    Amplitude drop-downs (not drop-outs -- see below) are still
present in some of the scans, with the same characteristics as
noted before. 

    Some additional information on these drops.  Careful viewing
of the behavior from the Saturday morning run show that the
the drops come in three flavors:

    a) Drop by 25% of the correlated flux (uncommon)
    b) Drop by 50% of the correlator flux (very common)
    c) Drop by 75% of the correlator flux (rare). 

    All observed drops, for the Saturday and today's data, are 
consistent with
this, to within the thermal noise.   As noted before, there is no
phase effect seen. 

    Special note:  This pattern is seen on the baseline 14 x 16,
independent of the levels of 14 and 16 with VLA antennas.  That
is, if 14 and 16 each drop by 50% as seen on baselines connected
to VLA antennas, the baseline 14 x 16 does *not* drop to 25% of
the full correlated flux.  It drops by 50%.  Thus, these drops do not 
'close'. 

    Ken notes that there are 8 waveguide cycles per 400 ms integration,
and has a conjecture to explain all of this.