[evlatests] general survey and referenced pointing

[Ken Sowinski]

Walter already reported on the L302 in antenna 24.  We
should try to get another pair of the latest L302 in
another antennaa to be sure that we can get it right
more than once.

I made a general survey this morning of what works:
X band: 13, 14, 16, 18, 24AC
C band: 13, 14, 16
L band: 13, 14, 16, 18
K band: 13, 14, 16        18 is broken
Q band: 13, 14, 16        18 is broken
P band  14                18 is broken

In spite of what I wrote about Q band last week, there
is no hint of any correlated signal for K or Q band at
antenna 18.  However there is no hint of anything wrong.
Synthesizers are locked; I even set up the second L301
to the same frequency as the first in case the wrong
one was connected to the receivers.  There seems to be
connectivity to the right receiver because when I turn
off the noise tube for the K band receiver the SD at the
backend goes away.  Looks to me like the UX converter
is doing something it ought not, or the feeds (both of them!)
are so far misaligned that the beam on the sky is a a degree
or more away from where it should be.  There is no correlated
signal in the crossed-hand correlation prodcuts either.

P band fringes are there, but I am not sure what we are
seeing.  There is lots of power in the crossed hands and
there is not much relation between attenuator settings,
sampler RMS and total power at the backend.  It is not
nearly as well-beahved as antenna 14.  Power meters in
the downconverters would help diagnose things.


The main problem with referenced pointing was uncovered.
Some combination of the change Bryan made concerning
when antennas register for pointing results and a change
I made sometime to telcal concerning when pointing results
are sent to the executor resulted in unreliable transmission
of pointing results as seen in the executor.  The simplest
explanatino of the symptoms is that sometimes for some
antennas results from one pointing scan earlier than
expected were used for pointing corrections.  I was able
to fix this with a change to telcal and Rick will exercise
it tonight.  In preparation for that I mad yet another
attempt to set interband collimations for K and Q bands.

In addition I am convinced that the az correction is not
divided by cos(el) before being applied, but this has not
yet been fixed.  The effect for low elevation sources will
be difficult to see if the pointing corrections are not
too large.