Ka-band observing guidelines
Compiled by A. Harris from Zpectrometer data and related email
exchanges with B. Mason and T. Minter
Elevation
effects
In short, observe between elevations of 20 and 80 degrees (below 75
degrees if
possible) for
reasonable telescope
performance, and at as high an elevation as possible for
atmospheric transmission considerations. Both pointing and
the
surface degrade at high elevations. The pointing
model is
most accurate from 20-80 degrees elevation, where data for the
model are taken. The surface correction tables have very few
observations at high elevations, and the surface may distort from
deformations in the backup structure, so the surface may deteriorate at
high elevation. The feed arm is vertical at 77.8 degrees, and
stresses in the backup support structure change rapidly at elevations
near this. A practical lower limit is 20 degrees elevation
for
reasons of gain, pointing, and atmospheric transmission (transmission
scales as t_z^AM, where t_z is the zenith transmission and AM is the
airmass, 1/sin(elevation)).
Wind effects
We measure 10% flux errors in repeated scans on pointing
sources
for winds of about 3 m/s. This is lower than the general GBT
recommendations but is consistent with observations reported
in http://wiki.gb.nrao.edu/pub/Dynamic/DynamicProjectNotes/dspn1.pdf
Observations seem to be profitable up to winds of 6-8 m/s.
Deviations from nice smooth pointing and especially
focus
curves are good ways to gauge the effect of wind on effective antenna
gain.
Surface distortion from
solar heating
The surface distorts markedly during rapid changes in ambient
temperature. Daytime observations can have aperture
efficiencies
below 70% of the nighttime values (a factor of two in observing time).
Daytime observations
should use the AutoOOF (automatic out-of-focus mapping to measure and
correct surface distortions); see http://wiki.gb.nrao.edu/bin/view/PTCS/AutoOOFInstructions
The suggestion is to run AutoOOF every hour or two; this is
insufficient during rapid temperature changes near dawn or sunset, when
we have easily measured varying
antenna gain on short timescales.
The wiki page suggests that a useful diagnostic is watching
for a
sidelobe in the elevation scans in the pointing curves (AutoPeak); a
noticeable sidelobe is a sign that an an AutoOOF is indicated.
Questions or comments? Please contact Andrew
Harris.