[evlatests] Notes from January 28 test meeting

[Barry Clark]

EVLA Test Meeting, January 28, 2010

1.  Widar current status.

Michael Rupen reports that things are starting to wake up.  There
are Station Boards for 27 antennas installed, with FORM modules.
Most of them work; there are a few known problems.  There are about
thirty Baseline Boards installed.  There a few with known problems.
(But note that only four are needed for OSRO.)  Electronics is still
working to get all antennas with optical levels set and in sync.
They claim that six antennas are working by their tests; in practice,
five of these give fringes, on one or more basebands.  Especially
considering the weather (site is closed for snow today), it will
take all of next week to get all the antennas up and fringing.

Software news:  the link from Executor to the Configuration Mapper
appears to work.  The remaining items before we can do an actual
end-to-end test are getting the Configuration Mapper to configure
the CBE, and providing MCAF with enough information about the
correlator setup to be able to write a correct SDM.

2.  Polarization.

R. Perley has been investigating the polarization properties of L, S,
and C bands.  The usual polarization methods ("D term determination")
measure the difference in polarization leakages relative to a mean
or reference antenna.  He has been investigating absolute polarizations.
The method employed is to rotate one receiver by 90 degrees.  For that
antenna, the mean linear component of that feed will show up in the
cross correlations with the unrotated antennas.  Presuming the rotated
receiver is typical, absolute polarization leakages can be measured.

There are six antennas with S band.  The absolute polarization leakages
run about 5%, which was the spec.  They get worse at the low end of the
band.

There are four Lband antennas with the wideband OMT.  Antenna 9 is
exceptionally good, with leakage about 1 % right across the band.  The
others run about 5%.

There are lots of wideband OMTs at C band.  The polarization leakages
run from about 2% up to 15%.  The newest receivers tend to be the best,
with leakages in the neighborhood of 5%.

3.  Event of January 10.

During the flux calibration run, at the very end of VLA observing,
an extraordinary event occurred.  From the beginning of the run until
0h IAT on January 10, lower sideband basebands came out phase
conjugated.  The problem corrected itself at 0h IAT.

There was a slightly non-standard setup for this run, in that not only
X band, which normally runs lower sideband, but the BD IFs of K, Ka, and
Q band were also run in lower sideband.

Whether other data than the first half of this run were affected is
unknown.

4.  Glossary
FORM   Fiber optic receiver module (also called deformatter).
IAT (or TAI)  International atomic time, currently 34 seconds different
        from UT.
MCAF   Metadata Capture and Formatter.  Records circumstances of an
        observation so correlations (stored elsewhere) can be related to
        source, frequency, bandwidth, etc.
OMT    Orthomode transition - extracts two polarization modes from a
        waveguide into coaxial format.
OSRO   Open shared risk observing - using Widar in one of two VLA
        emulation modes (but with more bandwidth and channels).
SDM    Science Data Model.  This is the metadata that MCAF captures.