[evlatests] Wobble dependencies

[Rick Perley]

    I took a closer look at the wobbles from last Friday.   The target 
source was 3C273 -- uncomfortably near the sun (perhaps 10 degrees).  
Ignoring the 2 and 4 Hz amplitude-only wobbles seen on Antenna 27, LCP, 
I find:

    1) The X and C band data were taken around 10AM local time, and show 
amplitude/phase wobbles only on baseline 27 x 28, in both RR and LL.  
The wobble frequency is proportional to observing frequency (and thus 
depend on sub-band frequency), and changes with time.   Only antennas 4, 
8, 15, 24, 27, and 28 were in the array at the time.  (NB antenna 19 was 
not present).  Antenna 27 is at E3, antenna 28 is at N8.   The other 
antennas are at:  4 @ W1, 8 at N1, 15 at W6 and 24 at W5. 
    The wobble amplitude is much higher at C band than at X-band. 
    In this configuration, with ~1 km maximum baselines, the maximum 
fringe frequency at 6cm wavelength is about 1 Hz.  The periodicity noted 
on 27  x 28 has a 2 Hz frequency -- at least 2 times too high, given 
that the sun (were it the source) is only a few degrees away from the 
fringe tracking center (3C273).  An exterior origin seems unlikely. 

    2) The L-band data were taken about 4 hours later in the day, with 
all 12 WIDAR-corrected antennas operating.  The wobbles are again 
visible, but not on 27 x 28!  The affected baselines (reported earlier, 
but largely restricted to antenna 19) were not present in the earlier X 
and C band data.  The fringe periods observed are typically 0.5 to 1.2 
Hz -- again a factor of ~two higher than would occur with an exterior 
origin.  Perhaps the best evidence against an exterior (sidereal) origin 
is that a 1 Hz wobble is seen on baseline 19 x 24 -- W4 x W5, a ~50 
meter baseline for which the fringe rate cannot exceed ~0.02 Hz.