The new data again show the recorded azimuths different by -0.01^o from the demanded azimuths. NPR thinks this is because the recorded azimuths are in the AZ system rather than the EN system of the demand : the difference being the IA term of the pointing model.

    Datasets        HST     mean leg temperatures   Humidity
                   start      start   middle   end       %
 ( 20001125   cw   11:55       2.4     3.6     4.8      25 )
   20001209   cw   10:55       2.5     3.6     4.5      30 

The azimuths were all incremented by 0.01 degrees for comparison with the last model, although the nnn.08^o data were then used to create the model that was to be installed. We shall adopt this convention, with much loss of accuracy, until further notice : 0.01^o is 1mm on the track.

The difference (in the 'nnn.09^o' sense) between the latest model and the previous one is shown below :

• There is a 3" spike in the difference data at approximate azimuth 142, coming from the older data.

• The formal differences in (F1,F2,F3) are described by the rms scatters of (0.32", 0.34", 0.33") cf (0.30", 0.21", 0.34") previously. The similarity in conditions prevalent during these two runs may result in the rather good, normal, residuals seen above.

• Installation of the new track model occurred at 17:30 HST 12 Dec 2000.

• The symmetry is fine - see the plot.

• The ripple from the right A-frame has diminished in amplitude rather abruptly since the last measures : see the plot.