[evlatests] P band, 4 band and power levels

[Ken Sowinski]

This is a more rambling than necessary summary of what was
learned last week by looking at sampler spectra and trying
to get P band to work.  The quick summary is that we need
to do a better job of matching spectral power density
between the two bands.


While I was debugging P band support last week Chuck was
able to capture sampler spectra for all the EVLA antennas
at P band.  Antennas 13, 16 and 19 were not working at 
P band at the time.  Antenna 24 may have worked, but its
L304 does not properly report band code which confuses the
spectrum gathering/plotting program.  Antenna 23 has no 
4 meter receiver package.

All IFs have spectra with similar features.  There is a broad
pedestal from about 25 MHz to 120 MHz on top of which is the 
4 MHz "spike" defined the the 74/4 filter in the T301.  P band
is visible as a 50 MHz passband from 300 MHz to 350 MHz.  The
noise floor is usually 20 to 25 dB below the 75 MHz pedestal
and the P band passband.  In many cases the 75 MHz pedestal
power density is the same or larger than the P band bandpass,
and because it is also twice as wide, in most T304s the total
power seen at the detectors is dominated by the 75 MHz noise.
This is a surprising result, but it is not obvious to me that
it matters.

At the VHF bands the T304 attenuators will be set the same for
both 4 and P bands and the FIR gain parameter in the deformatter
module used to match the 50 MHz output to the input power level
required by the VLA T5 module.  This will work only if there is
only a small difference in power density between the two passbands.
as seen in the T5.

In the first table below I list the measured power density for
each of the features, for all IFs to show the large variation
amongst IFs and antennas.  The most critical thing is that the
P band and 4 meter power densities be about the same in each IF, 
so the second table below shows the difference in dB between the 
power density in the 4 meter band and P band.  Differences between
IFs and antennas are unimportant.  Except for antenna 18 all the
4 meter bandpasses are about 15 dB above P band bandpasses.  Was
the T301 in antenna 18 built differently or is the solar attenuator
permanently inserted?


Table 1
        IFA        IFB        IFC        IFD
     Ped 4  P   Ped  4  P  Ped  4  P  Ped  4  P
14   25 42 24   25 46 25   24 37 31   20 33 30
17   30 42 28   25 36 21   26 37 30   25 34 25
18   26 38 35   19 29 27   31 42 35   20 31 28
21   25 43 26   17 34 18   26 43 29   19 34 20
23         37         32         36         33
24      ??         ??         ??         ??


Table 2
          PD_4 - PD_P
      A    B    C    D
14   18   21    6    3
16   14   15    7    9
18    3    2    7    3
21   17   16   14   14