[evlatests] More on Negative Switched Power ...

[Rick Perley]

    Ken ran a short test today, on all antennas and all bands, where he 
turned off the switched power so the negative offsets could be 
measured.  The antennas were stowed at the time -- this test simply 
establishes the presence or absence of this offset for the 'cold sky' 
situation. 

    The results show that the offset problem is present in six of the 
eight frequency bands.  However, some bands are much more seriously 
affected than others. 

    *  Bands for which no antennas show any offsets (at the cold sky 
levels) are:  L and Ku. 

    However, we do know from observations of Cygnus A, that there is a 
PDif reduction at L-band, so the absence of the offset for L-band noted 
in this test probably reflects the low power.  (I've asked Ken to repeat 
the test with the antennas pointed on Cygnus A). 

    *  X-band shows very few antennas with an offset, and the offsets 
are small.  (Antennas with offsets are 7, 17, 18, and 26). 

    *  S-band and C-band show the most antennas with offsets.  As 
always, antenna 24 in RCP, and antenna 1 in both polarizations, are the 
worst. 

    *  The three high frequency bands shows large offsets in about half 
of the antennas -- there is an interesting tendency for K and Ka bands 
to have far more offsets in LCP than in RCP. 

    For reference, note that the typical PDif (with the diodes 
switching) is between 100 and 400 'millicounts'.  A large offset seen in 
this test is about -10 millicounts.