41 pointing data were collected in good weather with RxA/ACSIS in a dedicated pointing run lasting about 3.5 hours. Removing collimation offsets of (CA,IE) = (-0.9,+1.8) gives the residuals shown below:

Raw rms residuals in (dS,dZ) are (3.9,3.6), about as before. The slight systematic in dZ-vs-Z seen in previous data is seen again (centre-left panel). This is removable - and removed - by the optimization of all 7 model parameters, as shown below:

None of the parameter changes is very significant, however:

* use ie ca ia npae aw an tf
* fit

       coeff       change     value    sigma

  1     IA         +2.091   -259.87    8.134
  2     IE         +2.048    -17.30    1.884
  3     NPAE      +16.770    +39.03    9.720
  4     CA        -14.955   -172.91   11.866
  5     AN         +0.626     +8.99    0.818
  6     AW         -0.619    -15.76    0.765
  7     TF         +0.338     -3.92    3.070
  8     AUX1S                 +1.00
  9     AUX2E                 +1.00

Sky RMS =   4.47
Popn SD =   4.91

with the larger changes - to NPAE and CA - probably being due to their strong correlation. The rms residuals reduce only to (2.8,3.5).

The persistently poor-ish azimuth performance might once have been ascribed to chopping problems or to a substantial change in the central bearing race defect - a check of which is still pending. However, the SMU was reported to have behaved fine throughout, and the plot below would refute a claim that the azimuth residuals are correlated with the phase of the central bearing race defect:

The v.poor elevation performance seems to indicate some other, more pervasive, source of noise. (The good weather was important in eliminating atmospherics from the mix).

I have also enquired as to when the inclinometry system will be restored; although the plot below show little correlation between the pointing residuals and track joints (shown as the blue lines) :

The new model was installed at 18:00 HST on Jan 13.