Inclinometry was taken for the first time with the new TCS

    Datasets        HST     mean leg temperatures   Humidity
                   start      start   middle   end       %
 ( 20000713_3 cw   02:45       6.8     6.2     5.9      70 ) 
   20000922   cw   10:15       3.4     4.7     6.0      30

The difference between these models is shown below :

• The new dataset contains electronic spikes at azimuths 43, 361 and 373 - these are removed before the creation of the new model.

• With these spikes removed, the formal differences in (F1,F2,F3) are described by the rms scatters of (0.69", 0.64", 0.82") cf (0.52", 0.26", 0.41") previously.

• The new data show a strong ripple of frequency ~8.8 degrees. Analysis of the raw data show it to arise from the right side of the antenna. Previous manifestations of this ripple have been interpreted (successfully in at least one case) as a diagnostic of a broken wheel, although subsequent analysis offered an alternative interpretation - that of slippage of a (permanently) defective wheel generating a phase error that manifests as a ripple. Slippage could occur if the load changes - and both the amplitude (about 1" peak to peak, as last time) and the symmetry data below tend to support this hypothesis in this case also.

  However, some precautionary inspection will be performed soon. Installation of the spare inclinometer on the 'radial' arms (beams) going from the central bearing towards the wheels will allow unambiguous identification of the wheel involved, although it is already narrowed down to the right side and the rear-right wheel was the prime suspect last time around. Fourier analysis of recent data should show when the damage or slippage occurred.

• Installation of the new track model is pending.

• The symmetry has remained less than optimal since the central bearing load adjustment on 06 June 2000 - see the plot - but is currently worse than it has been in about 2 years. This may be related to the new wheel problem, but in any case may reflect a separate need for a new c.b. load adjustment.

2000 09 27
The Fourier analysis program (/home/imc/pointing/progs/incft.exe) was re-written to use the PDA Fourier algorithms (rather than those from the NAG library, to which we now no longer have updates or access). Analysis of the difference data below reveals low power even in older data at the 8.8 deg. frequency expected from a single revolution of one of the wheels, but an order of magnitude more power and general overall consistency in the most recent difference data :

Only the Left A-frame Y axis inclinometer sees none of the 8.8 degree ripple, suggesting the problem lies on the right side of the antenna, as suggested above. Inspection of the wheels is underway.

2000 10 03
A meeting was held to discuss these results and concluded that the signature is of a wheel misalignment rather than defect - link here. Fourier transform analysis of inclinometry data will provide a useful method of monitoring the problem.