SUMMARY
Further tests of the 's-correction' and some all-sky pointing were done over the past two nights with the new TCS.

• The new data show again that the step is not symmetric about azimuth 180 degrees, in stark contrast to the discovery data.
• The empirical correction is being added with the wrong sign within the new TCS.
• All-sky pointing allowed a new TPOINT model to be generated.

Transit Tracking
The 'transit' problem was noticed first in April . A correction was installed on 16 May, but recent checks in May, June and twice in July have been inconclusive for various reasons.

The latest tests were done during the last stages of commissioning of the new TCS (Telescope Control System). Conditions of moderate transparency (tau_cso=0.10) prevailed. Uranus was tracked for about 2 hours on consecutive evenings, using the SCUBA map16 method; the first time with the EMPIRICAL correction de-activated, the second with it activated :

Transit occurred at elevation 54 degrees. The raw residuals are (1.3",1.4") in (daz,del) on the first night when the seeing was 0.5", and (0.9",2.7") on the second, when the seeing was 0.8". The del residuals clearly show the 4" step in the first dataset, although the step is not symmetric about azimuth 180 degrees, as was the case during the discovery data. Similar 'transit-asymmetric' residuals were noted in the previous runs, although these new data suggest a symmetry point only 1 or 2 degrees from 180 rather than the several degrees seen previously.

The second night's data appear to have added the empirical corrections in with the wrong sign : the step ought to have been reduced to zero, but appears instead to have doubled in amplitude.

All-sky pointing
19 data were collected before the first of the transit experiments above. The raw residuals are shown below :

The data show systematic errors in the daz-vs-az and del-vs-az plots indicative of errors in the values of the parameters describing the deviations of the azimuth axis from the vertical. The data were run through TPOINT and the raw residuals of (3.4",3.3") are reduced to a vector error of 2.3" after the following changes to the TPOINT parameters :

          Parameter    AN     AW   NPAE     CA    IA    IE    TF
   Suggested change   3.1   -2.0   -1.6   -1.0   0.2  11.6   9.8
                 +-   0.6    0.7    2.2    0.3   2.2   3.2   4.5
          New value  11.3  -18.2  104.6 -222.9  31.8  77.1  -7.4

It is noteworthy that TPOINT did better at providing a controlled solution for these data than FIT9, which, as often happens, allows parameters 3,4,5 (NPAE, CA, IA) to change by self-consistent but excessively large amounts. TPOINT is to be used exclusively once the new TCS is adopted.

Conclusions

• The new transit tracking data suggest that the s-correction is once again a fair approximation to the form of the elevation error, although the characteristics of the error appear to have varied since the discovery in April, principally in the azimuth of symmetry.
• Several of us (WSH, PF, ENA, NEJ, IMC) met Friday afternoon and suggested that the next tracking data be taken using RxA and the repetitive fives routine. This will eliminate SCUBA, residual clock errors and my customized reduction script from the process.
• The adjustments to the TPOINT model were made during the afternoon 25 Aug 2000.