[evlatests] Fooling 'set-and-remember' at L-band

[Rick Perley]

    I wrote earlier about the Psum values being low (~factor of 10) for 
three antennas (ea02, ea07 and ea13) in RCP on a recent test.

    An almost identical test was run the previous day.    Looking at 
this, it is immediately apparent that the problem noted above was *not* 
present in the earlier test.  All PSum values, for both IFs, for all 
three antennas, are well within the accepted norms.

    I claim that 'set and remember' got fooled -- somehow, but an 
erroneous high IF power during the 'set and remember' stage of the 
observation.  But why only those antennas, on that one polarization?

    We've seen this problem before, and I'll again point out that the 
receivers are more stable than the L-band power environment.  (Same is 
true at S-band).  I know there is great resistance to utilizing pre-set 
tables.  But it seems to me that table errors of an order of magnitude 
are much less likely than the errors in the 'open-sky' power settings 
that we're occasionally seeing with the current arrangement.