[evlatests] Switched Power Compression and Offsets at L-band

[Rick Perley]

    Ken ran a script yesterday which:

    1) Observed Cygnus A with the switched power monitoring system both 
on and off; and

    2) Observed nearby blank sky, with the switched power monitoring 
system both on and off. 

    This was done at the four lowest frequency bands:  L, S, C, and X. 

    This note summarizes the results from the L-band observations, taken 
in two IFs, with AC at 1394, and BD at 1746 MHz.  At this frequency 
Cygnus A approximately quadruples the total system power -- this is well 
within the linearity specifications, so we expect the PSum values to 
show this ratio, and the PDif values to remain unchanged between the 
Cygnus A and cold sky observations (when the switched power injection is 
on).  

    Major Results:

    A)  There are no 'zero offsets' detected when the switched power is 
turned off!  This is an interesting result, as these offsets are 
repeatedly observed at C and S bands  (see later notes). 

    B) Nearly all of our antennas show significant compression 
(reduction) in the PDif value when we observe Cygnus A, compared to the 
nearby cold sky position.  The typical compression is 10 to 30%.   
Previous observations have demonstrated that this compression is not 
seen in the visibilities -- it is apparently a problem with the 
monitoring system itself. 

    C) A few antennas show *no compression at all*:  ea11 in all four 
IFs, ea17 in all four IFs, ea28 in LCP, and ea14 in RCP.  These are then 
the 'golden antennas' -- the goal would be to make the others like 
these!  A few other antennas, like ea02, show a compression of less than 
5%. 

    D) Conversely, there are a few antennas with much larger PDif 
compression:  ea25 in RCP has a nearly 50% reduction in PDif when on 
Cygnus A. 

    Some Minor Results:

    A) Antenna 24 showed no power at all (i.e., the SY tables were 
empty) in IF1.  IF2 was present and normal. 

    B) Some antennas/IFs had a very low power level:  Ken tells me that 
we should be seeing a value for the power of about 15 'counts' when on 
cold sky.  The majority of the antennas in IF2 showed values close to 
this.  The average power level in IF1 was about 10 counts -- 50% too 
low!  And some antennas are nowhere near correct: 
     ea01 in IF 'B':  power level at 4.1, and the ratio of Cyg/ColdSky = 
1.5, rather than 3 to 4. 
    ea03 in IF 'D':  power level 3 times too high. 
    ea10 in IF 'A':  power levels a factor of 5 below the average for 
IF1 (and a factor of 8 below IF2). 
    ea15 in IF 'A':  low by a factor of a 3 to 5 in all four IFs.
    ea28 in IF 'A':  same as ea10.