Knowledge of the accuracy and repeatability of the inclinometry system is essential if any of the more subtle behaviours of the antenna and track are to be quantified. To this end many inclinometry runs include repeated clockwise (CW) or counter-clockwise (CCW) datasets. A comparison of CW or CCW inclinometry runs taken in quick succession on the same day (950609) is described in the table below. The RMS differences between two CW runs (CW1 & CW2) are shown :
Ch 1 2 4 5 7 8
TX TY LX LY RX RY
CW1 - CW2 0.21 0.35 0.76 0.26 0.48 0.28 " rms
The 4 channels currently employed by the inclinometry reduction algorithms are
Channels 1, 2, 5 & 8. Channel 2 (PITCH) directly yields the elevation pointing
error, while the other 3 channels combine to yield ROLL and YAW and hence the
azimuth pointing error. Thus the inclinometry errors contribute a maximum of
0.40
of error to azimuth, more typically 0.30
at
intermediate
elevations, and 0.35
to the elevation error. We adopt
0.35
as
the typical inclinometry noise level.
in Channel 8 (Fig.4).
The step occurred at one of the largest track joints (wheel 2 at joint 14/1)
and could be either a shock-induced zero-point change in the inclinometer,
or a real change in the antenna structure as a result of the same shock.
(see the report of 950523).
This has implications for the systematics occasionally observed in the
Overlap region and described below, but also illustrate the compelling
need to remove the large bumps in the track for more than just aesthetic
reasons.