[evlatests] Phase jump, and more

[Rick Perley]

    A quick stress test was run.  Early results:

    1) There was a global EVLA to VLA phase jump, at L-band, of 100 
degrees, between the first (so-called 'dummy') scan and the second 
scan.  This occurred on IF #1 *only*.  I added two more L-band scans, to 
follow the two listed above -- no jumps occurred on these.  The jump 
noted is exactly the same as noted in earlier tests by Vivek and me.  It 
smells like a software-induced thingy. 

    2) A check of the frequency axis at P-band was done -- all is 
correct, in that the baseband phase 'hook' is on the left side, as it 
should be.  OTOH, Aaron Cohen sent me his 4-band BP solutions, from his 
Sept 1 observations, and the 'hook' is clearly on the *right-hand* 
(i.e., wrong) side for the EVLA antennas.  More investigations will be 
needed tomorrow. 

    3) The 'fast dump' solar mode was not working -- we got 3.33 second 
averaging, not the 0.418 seconds requested.  Bryan has already uncovered 
the origin of this.