The Point-Wise data OUTCW1, OUTCCW1 and CW25 are sufficiently similar that
the mean of the two CW runs (OUTCW1 and CW25) were used to create CW940925
(see above) that represents our most reliable CW data and our confidence that
it reflects the CCW data bar random errors.
This dataset has been used to create the track model MODEL1-940925 which has
been used since that date. This model is shown in Fig.7. As reported on
941018
the performance of this model, and the accompanying 7-parameter antenna model,
has been confused because of problems caused by the September heavy
engineering. However, it has become apparent that the elevation pointing (from
all-sky data) has improved considerably, from rms errors of about
2.0" to about 1.4". This is probably because the TMU-Y axis
measures the elevation pointing correction directly and is now simply doing it
better than previously. The lack of associated and clear improvement in the
azimuth pointing is perhaps because the theory of conversion from A-frame
inclinometry to yaw is flawed (see above). Early tracking experiments (on
Saturn, from (az,el) = (163,58) to (180,59), passing over 4 track joints) gave
rms errors of (0.3", 0.3") (16 points logged) - which is the
encoder resolution limit !!