[evlatests] Strange Things ... (cont.)

[Rick Perley]

    Michael took lots more data yesterday afternoon, which I am trudging 
through.

    The databases under consideration now have X and C bands alternating 
for 45 minutes, one minute at each band. 

    I've analyzed the X-band one, which has some of the usual phenomena, 
some things missing, and one new one.   Read on ...

    1) Antenna 2 gave no fringes throughout.  (It was also not fringing 
at C-band). 

    2) Three baseline/IFs gave identically zero amplitudes throughout -- 
these are the same as those reported on yesterday.  They are:

       25 x 19 and 25 x 24 on IF #1 (sub-band 0)
       25 x 24 on IF#2. 

   3)  The delays are the same as reported yesterday. 

    4) There are *NO* 180 degree phase flips.  In fact, there are no 
phase jumps at all. 

    5) Numerous 1 dB amplitude changes were observed on 5 of the 
antennas:  1, 3, 18, 24, and 28.  Each of these has its own pattern -- 
there seems to be no correlation between them.  See below for BP changes ...

    6) The first scan (19:25:08 through 19:26:05) had completely 
different amplitudes than any of the following.  The changes between 
this scan and the others is unrelated to the 1 dB changes seen afterwards. 

    7) Bandpass solutions were made for each 1 minute scan.  The first 
scan (with the weird amplitudes) also gave a weird bandpass solutions, 
with long-wavelength oscillations.  All other scans were stable, and *** 
there is no effect seen, to the <1% level* of any change in bandpass 
when the 1 dB amplitude change was in, or out ***. 

    8) A unique new phenomenon was found:  Baseline 23 x 25 has an 
exactly two turn phase wrap across it -- completely independent of all 
other antennas and baselines.  This is seen in all IFs.  This is a 
non-calibratable effect (via existing software). 

    9) Following bandpass calibration, and concatenation to a single IF 
(of 925 channels), the amplitudes and phases were investigated for the 
standard oscillatory phenomena.  There are there:

       Baseline 19 x 25.  Amplitude period 15 second, phase period 6 
seconds, phase amplitude 1.5 degrees. 
       Baseline 19 x 23.  Amplitude and phase periods both 3 seconds.  
Amplitude amplitude 5%, Phase amplitude 2 degrees. 

    No other baseline shows any unusual excursions.  This doesn't mean 
the oscillations are not there, only that either they are so fast they 
are averaged out in the 1second integrations, or they are low enough 
that I can't see them in the noise. 

    I'll review the C-band data next.