[evlatests] Two more notes on PDif and Switch problems

[Rick Perley]

     1) Plotting the PDif values vs. elevation, over a period where the 
temperatures were relatively stable easily showed the expected variation 
with system temperature.  The effect is best seen at 48 GHz, where the 
Tsys doubles between zenith and 10 degrees elevation.  For those 
antennas that do not have the switch problem, the compression of PDif is 
typically 5%.  As usual, some antennas show it more strongly than 
others, and there is no correlation between polarizations.  This is 
consistent with the low frequency observations of Cygnus A.

     2) In viewing the PDif values for the T303 switch issue more 
carefully, I see that by far the most common problem is that the value 
is off (meaning, the gain has been reduced) by 5 -- 20% -- and is 
generally stable during the (short) observation.  So, while some of the 
PDif values are changing within a scan, these are actually in the minority.