[evlatests] L-Band Sidelobes, and Cygnus A

[Rick Perley]

    WIDAR permits us to see things that we always knew had to exist, but 
could never actually observe ...

    I've noted earlier that the L-band data from our Cyg A test seems to 
be afflicted with wideband interference.  This statement was based on 
quick perusal of the calibrator data, which has many deviant amplitudes 
and phases, seen both in the VLA correlator and WIDAR correlator data, 
with both VLA and EVLA antennas. 

    The calibrator, 2007+404, is only 1.6 degrees away from Cyg A, and 
is about 500 times stronger (~25 dB).  The L-band primary beam's forward 
gain provides about this much attenuation at that offset -- depending 
very much on whether one hits a sidelobe or a null. 

    Using POSSM, one can see the beating between Cyg A and 2007+404, 
when the antennas are pointed at the latter.  The beat is seen on all 
baselines with varying period (as a function of frequency), and is 
*very* much stronger on sub-band 1 (centered at 1308 MHz) than at any 
other.  Sub-band 3 (1820 MHz) is by far the least affected -- 
apparently, at that frequency, the beam has a null at a radius of 1.6 
degrees.   The period is quite compatible with the hypothesis of Cyg A 
beating with the calibrator.  The amplitude of the beat is as great as 
10 dB between maximum and minimum -- but usually rather less. 

    Conclusion:  Don't observe near Cygnus A if you want to make 
noise-limited images ...