The temperature parameters of the JCMT pointing model

Besides the 7-parameter model describing the imperfections of the antenna, the track model determined by inclinometry , and the refraction model , there are a couple of recognized ways in which the temperature of the antenna impacts the elevation pointing. As a result, there are two parameters in the file THI.tide that need consideration when the elevation pointing goes awry :

• temp_slope relates del to (T_f-T_b), where T_f is the temperature of the front legs of the antenna, and T_b is that of the back legs. temp_slope currently has a value of 6"/^oC, in agreement with simple engineering calculations.

• temp_mean_slope relates del to 0.5*(T_f+T_b), and currently has a value of +1.0"/^oC.

indicates typical diurnal variations. The positions of the 8 probes are included in the plot (lrf = lower right front, etc), with the convention that the arrangement is as viewed from behind the Antenna Engineering Console (AEC) looking out towards the main doors.

These data are combined as described above to give, for instance, T_f-T_b, which varies diurnally as seen here (over a couple of days immediately after the recalibration on 06 Aug 2004):

Impact upon pointing
Solar infrared illumination (insolation) of the front legs through the membrane will cause T_f-T_b to increase. The front legs expand, the dish tips backwards and the antenna must be moved to lower elevations than expected to acquire the source. The pointing error del, or uel as it is seen on the DISPLAY screen, is then negative. The T_f-T_b differential can be particularly strong immediately after sun-up:

with the possibility of a 2degC (=12") change. Plots of pointing data occasionally show del varying systematically throughout the evening shift, and a plot against T_f-T_b may indicate an error in the current value of temp_slope. However, elevation pointing immediately after opening the doors may be affected by several things, and it may take the combination of several nights worth of data, covering a large range of temperatures, to be convinced of the need to change the value. (Also be careful of the sign convention).

In the past, some strong variations of del with time (i.e. (?) temperature) have been found to be immune to changes to temp_slope, and it was necessary to worry about other possible thermal effects. The temp_mean_slope parameter was introduced on 30 Apr 1997 to allow corrections that may be functions of the mean antenna temperature. Both parameters may also be seasonally variable.