SUMMARY
The new data are part of an attempt to reveal the extent and size of the elevation pointing errors induced by the reversal of direction in elevation that becomes so apparent during transit tracking (see the interim summary report of 20000408 and the problems page on the web).

The new tracking data at elevation 40 degs confirm the extent and amplitude of the problem see previously at elevation 70 degs. The sense of the elevation pointing error following a slew is convincingly a function of the direction of the slew in elevation space, and the relaxation afterwards is seen to occur over about 10 miuntes.

Data was collected with SCUBA in its conventional POINTING mode during the first part of 1st shift (HST 18:30 to 21:30). There is the possibility of considerable thermal impact on the elevation pointing which will hamper the analysis, but corrections for these impacts are described below. POINTINGs were executed on a handful of sources in order to test the effects of cis- and trans-meridional slewing, and by lingering on each source for up to 30minutes the subsequent behaviour of the antenna can be compared with the mechanical explanations of the release of wound-up energy. The sources, elapsed times, azimuth and elevation ranges covered and the directions of slews etc. are listed below :

  Phase      HST       source        az      el        Notes 
  -----  -----------  ---------   -------   -----    --------------------
    A    18:45-19:00  irc+10216     103      67      Point & focus
    B    19:00-19:30  1034-293    147-154   32-35    baseline tracking
    C    19:30-20:00  0607-157    233-237   34-30    E-W slew
    D    20:00-20:30  0736+017    239-245   58-53    cis slew in W
    E    20:30-20:50  1124-186    149-153   45-47    W-E slew
    F    20:50-21:40  1034-293    173-188   40-40    cis slew in E +
                                                     transit at el=40

The total raw pointing residuals are shown below left, while the behaviour of the azimuth and elevation residuals in the azimuth plane are shown below right :

The raw rms residual in azimuth is a fairly impressive 1.3", while the plot of del -vs- time (left frame, centre-right panel) shows a significant increase during these 3 hours that is probably thermal in origin. Attempts below to monitor gradual drifts of the elevation pointing residuals must be tempered by this possibility. The two most common dependencies are of del with (T_f-T_b) , the difference between the temperatures of the front and back legs of the antenna, and of del with 0.5*(T_f+T_b) , the mean leg temperature. As it turns out, the observed changes in elevation pointing tabulated below seem independent of any such gross thermal correction.

The behaviour of del during the phases above is tabulated below to an accuracy of 0.5" :

  Phase    Description       Change       Immediate   Subsequent
                          in elevation      change      changes
  -----  --------------   ------------   ---------   ----------
   A-B    CIS slew in E       -35           +4.0       -2.0       
   B-C    E-W slew              0           +1.5       -1.0,+1.0
   C-D    CIS slew in W       +30           -2.0       +1.5
   D-E    W-E slew             -7           +3.0       -1.0
   E-F    CIS slew in E        -6           +1.5         -
    F     transit in S         ..             -        -2.0,+5.0      

Note

• The recorded seeing decreased from about 1.0" during phase B to 0.2" by the end of phase F. Phase B was designed to measure the nominal tracking capability, but was the worst affected by thermal problems. The rms scatter of del during phase B was 0.7". Hence the accuracy of the entries in the table above.

Conclusions

1. The correlation of the immediate pointing change with the change in elevation is clear, although the size of the effect seems uncorrelated with the elevation distance travelled. One might arrive at a 2.5" average : -ve for slews up, +ve for slews down. This could be the subject of further research or experiment - and I probably have stuff from Remo or from my own analysis of dedicated pointing runs to link to from here.

2. Subsequent changes in del, i.e. over periods of ~10^m, also seem correlated with the direction of elevation slew, with some waffling when there is essentially no change in elevation. An average value of 1.5" might seem appropriate for a 10^m interval; in the opposite sense to the immediate change and consistent with the recovery of windup energy .

3. There seems to be no correlation between changes in del and E-W, W-E or either type of CIS slew per se.

4. With an arbitrary adjustment in zero-point, the transit data are seen to be in rough agreement with previous southern transit tracking data :
   

   - although the first few data are dominated by the recovery phase from the previous elevation slew, and the full amplitude reached appears larger than in the previous data taken at higher elevations (~70^o). The amplitude of the error across transit of roughly twice that seen in a simple elevation slew has also been explained mechanically .