Updated telescope model (again)

The telescope model I installed yesterday did not produce the immaculate performance I was expecting :

In fact, the behaviour of the azimuth residual with elevation is the mirror image of that seen yesterday , although of lower amplitude. "Why should this be ?", I hear you ask ! Well - I don't know. But I must add that, at least during 2nd shift, local pointing was always done and so these trends do not impact the quality of the astronomical data taken around these pointing measures.

Once again, the elevation residuals are a strong function of time and this needs removal before the adjustment to the telescope model is calculated.

The strongest trend is

       del  =   -4.5 * (Tf - Tb)   + const
              +- 0.8

   where Tf and Tb are the mean temperatures of the front and back legs. 

The resulting (temperature-corrected) data were then run through the FIT9 program to deduce changes to the 7 parameters of the telescope model that minimize the residuals. The following changes :

       parameter #   1     2     3      4      5     6     7
                  -0.7  -0.1   1.8     6.4    6.9   1.2  -2.5 
 (c.f. yesterdays) 0.2   0.1  -7.3   -14.9  -12.7   0.0   0.1  

gave the following pointing residuals :

The pointing model has been updated and should now produce rms scatters in (daz,del) of (1.3", 1.2").