[evlatests] Sub-band saturation and RFI

[Todd R. Hunter]

On Tue, 18 May 2010, Rick Perley wrote:

>    The basic cause of this first-seen phenomenon seems fairly clear:
>
>    A strong source of RCP radiation (almost certainly a GPS satellite,
> but this is unproven) passed close to the boresight of the antennas.

The offender was GPS satellite Navstar22 (USA66).  I downloaded the TLE's 
for all current 25 GPS satellites from my space-track.org account, and 
used the TLE-to-az/el converter software that I assembled for GBT 
holography sessions.  Here is a graphical result showing the 2-hour path 
for all satellites above the horizon, followed by a zoom of the offender 
path in blue and the EVLA L-band beam location in red:

https://staff.nrao.edu/wiki/pub/Main/GPSSatelliteEncounters/gps.png

The duration of the arrows correspond to the time range of the effect 
reported by Rick.  Due to the high elevation of this encounter, I have 
ignored refraction. Given an accurate (az,el) time series for the 
antennas, I could produce a more exact plot of angular separation vs time. 
The TLE's are generally good to within about 2 arcmin for well-behaved 
geostationary satellites, but I don't have any experience observing GPS 
satellites. Presumably they are well characterized.

As an aside, the *elevation* values that Rick reports from AIPS agree with 
the values from the monitor data query tool for this time period, however 
the *azimuth* values do not.
    https://mcmonitor.evla.nrao.edu/monartool-gui/main.faces

I have accounted for the fact that the monitor query is expressed in local 
time (-6h) rather than IAT.  A closer look suggests that the monitor data 
query result may be wrong by exactly -180 degrees (it reads 134 instead of 
314). I then did a spot check of the current observing position with the 
monitor query tool and found that it currently agrees with the live values 
on the Multicast channels (117 deg).  So this is another mystery worth 
investigating.

Todd