[evlatests] Switched Power Issues

[Rick Perley]

    There are a small number of significant issues with the switched 
power system that need to be addressed.   From the L, S, C, and X-band 
observations last night, I find:

    1) The PDif values from ea28, in LCP at L-band provide no useful 
information -- although the values are at the levels expected (a couple 
hundred 'millicounts'), they contain no information -- only noise.  The 
PSum values, although clearly seeing the sky, are also dominated by 
noise and/or large gain variations.  This only occurs at L-band, in the 
LCP polarizations.  Other bands are fine for this antenna.    This 
problem has been reported many times -- can nothing be done? 

    2) The PDif values at X-band from ea05 on all IFs are essentially 
zero -- about 5 millicounts, when they should be 500.  The PSum values 
look fine.   It acts like the noise diode is not switching. 

    3) The PDif values from ea12, in IF 'A' only, are unsteady, and are 
often a factor of 10 or more too high.  They are also of the appropriate 
level at times.  PSum behaves in a similar way.  The visibilities also 
behave badly, but the application of the PDif does not fully recover the 
proper amplitudes.  This occurs at all four bands observed, but only on 
this single IF.   It acts like an attenuator was not set correctly at 
the beginning of each observation -- but did eventually get around to a 
proper setting after 30 seconds or more had elapsed.