Report to the JCMT Board - October 1998
Report to the JCMT Board - October 1998
Central bearing defect
The empirical correction for the central bearing problem remains
in effect, and measures of the effect upon pointing are made rougly
each 2 weeks. The measured peak-to-peak amplitude (at the horizon) is seen
to vary between 10
and 16 arcseconds and may correlate with time elapsed since the
(infrequent) central bearing load adjustments. Occasional adjustments
to the empirical correction have been implemented in order to minimize
azimuth pointing errors. The replacement of the bearing is scheduled for
May 1999.
SCUBA mapping mode as optimal pointing configuration
The above tracking measures are best facilitated by use of SCUBA 16-point
(850um) jiggle maps. Each integration within such a map yields a centroid
to give the pointing offsets (and the associated time, azimuth and
elevation) with a time resolution of 36 seconds. This is about 4 times
higher resolution than standard SCUBA or heterodyne (5-point) pointing,
and has proven optimal for describing the central bearing defect profile.
8.8 degree periodicity in inclinometry
In the early part of the Semester 98A measures of the track
irregularities (using tilt meters strategically placed on the antenna)
showed a component with a periodicity corresponding to
the rotation of a wheel. Previous detections of this periodicity
caused concern that this was a precursor to a wheel failing
catastrophically. However, further analysis showed the amplitude
to have risen and fallen over the previous 1-2 years,
and that it could also be explained by a change in phase of a wheel
with slight ellipticity and was not necessarily a harbinger of doom.
The phenomenon is monitored weekly.
Erratic variability in measured track profile
More recent measures of the track profile have also shown erratic
variability of order several arcseconds that still need explanation.
The profile measured is 'dynamic' in the sense that it is a function
of loading by the antenna, and the variability may imply variable antenna
loading plus a recent increased sensitivity to such. This may be due to
a non-optimal distribution of loading on the wheels, which, in turn,
may be a feature of the central bearing problem, above.
Tests are ongoing to resolve this problem.
Summary pointing accuracy
As a result of these uncertainties there are many caveats to any single
number describing the current pointing accuracy. One would like to
believe that an underlying inherent accuracy of something
better than (1.5",1.5") rms in each
coordinate persists, but central bearing errors
of up to 3"*cos(elevation) may affect 2 degrees of azimuth in every 16
at some times, and a similar quantity with a more complex
(track-profile related) azimuth distribution results from the
track-profile problems. The latter problem also causes smaller errors,
with similar azimuth distribution, in elevation pointing.
Return to POINTING
Iain Coulson
01 Oct 1998
|
