The effect of the "hump" in A3's receiver temperature and the assumption of sideband gain equality was studied extensively during commissioning in December 1998, partly as an outcome of observations of SO and SO₂ at the time, and the results are reported here.

In A3 the integrated line intensity ratio of a line observed in the USB versus the same line observed in the LSB shows some considerable variation with frequency, from 3 - 7% greater for CO 2-1 at 230 GHz up to 53% greater for SO at 258 GHz. These differences correlate with the receiver temperature values Trx, as shown in the plot below.

For a given line, say SO at 258 GHz, the receiver will be tuned in the USB at an LO frequency of 254 GHz where Trx = 154K and then in LSB at 262 GHz where Trx = 66K. In this case, the Trx changes are due to the conversion loss changes of the mixer: the mixer is relatively more or less efficient at coupling the input radiation. In the case of the CO 2-1 line, the mixer behaves the same way in the upper and lower sidebands, resulting in very small changes, around 5%, in the measured line strengths.

Thanks to Stephane Claude, NRC/HIA, Victoria (now at IRAM), for providing the above information.