[evlatests] EVLA Tests, Dec 9, evening

[Mark Claussen]

On Friday evening, Dec 09, we ran a test file to check that changing
the LO would alleviate the suspected aliasing problem.  Ken had checked  
this briefly during software time on Friday and found that indeed changing 
the first LO did indeed make the "C band look much cleaner and apparently
increased G/T."   Walter created a file that stepped the 1st LO from 
13184 MHz to 14976 MHz in steps of 256 MHz for both IF pairs, for C band, 
and from 13184 to 16256 for K band.  The integration time was 0.4 seconds.  

In the second part of our test we did the same thing except for observing
in spectral line mode (mode 4, 6.25 MHz bandwidth) in order to try to check
the sensitivity.  Here the integration time was 10s since we were doing
line mode.


Here is a quick summary:

1.  For the continuum scans at C band using 13952, 14208, and 14464 MHz there 
were no fringes EVLA-VLA baselines.  There were fringes 14-16, but with winding 
phases.  For the scans with other L301 settings, fringes were seen on EVLA-VLA 
baselines.  After closer examination, the phase windings do *not* occur at scan 
changes, i.e they do *not* occur exactly at a change in L301 setting.  All four 
IFs participated in this behavior.


2.  There were also no fringes on EVLA-VLA baselines at all during all the
scans at K band.  On the 14-16 baseline there were fringes, but again, with
winding fringe phases.  Again, all four IFs had similar behavior.

3.  During the line observations at C and K band, there were, again, no fringes
on VLA-EVLA baselines, with similar phase winding.  During the K band line 
observations, the fringe-winding went through a flat period (similar to a source
or antenna position error).

4.  The original intent of the observations, to confirm if a different LO setting 
would eliminate or ameliorate the aliasing effect, was acheived for C band.
For K band, this was hampered by there being no fringes at all on EVLA-VLA 
baselines (for comparison), and by the lower signal-to-noise ratio for K-band
observations.  So this intent was not acheived at K band.

5.  I also concur with Ken S. that 16A seems to be less sensitive (lower 
amplitudes) than other IFs at C band.


I will post details of the phase winding in a separate message.

Mark