Amplitude Calibration at Tr
during EVN Session 2/2009

What is striking in this example is that out of 5 cycles of Tcal measurements only one (that marked with the blue digit 4) went as expected! In the cycle numbered 1 we see that for the last two measurements the antenna did not return on the observed source until after the last measurement (ONSC) of the cycle. Similar was the case in cycles 2, 3 and 5, where during each of the second ONSO/ONSC pair the antenna only started to move to the source. In addition, in cycle 2 the OFFC is strongly affected by improper timing.

The table below shows results of calculations (for lower sideband of BBC 1 only) of the diode temperature Tcal for each of the five cycles as given in the log file, Tcal(FS) = Tcal(j)*DPFU (same is used by GNPLT application), and the variance weighted values, Tcal(VW). As could be foreseen, the two Tcal agree only in the cycle 4, elsewhere Tcal(FS) being strongly overestimated. Note that RCP was measured only in two of the 5 cycles.

Tcal in BBC1l calculated by Field System and that based on weighting by reciprocals of the variance Cycle 1 2 3 4 5 Tcal(FS) [K] 21.5 17.1 23.3 9.26 21.6 Tcal(VW) [K] 10.6 7.3 11.5 9.28 10.6 Polarization LCP RCP LCP RCP LCP Freq [MHz] 4964 4964 4896 4896 4960

Although this example presents rather extreme case, nevertheless it illustrates well how serious is this problem at our station. That is also the prime reason why we cannot use the otherwise very useful GNPLT program. Fortunately, we have a completely new, and practically ready, antenna control software which should replace the current faulty one really very soon.