Sensitivity (small changes in numbers from last messsage) --------------------------------------------------------- Guideline for computing integration times have been provided by Volker Ossenkopf from the ICC : - Use SSB system temperatures T_sys everywhere and state the values used - The efficiency of the observations will be about half that of the corresponding "ideal mode". T_sys : system temperature (K) delta_nu : frequency resolution (Hz) T_tot : total observing time (s) A : ``efficiency factor'' Depending on observing modes, this efficiency factor is : Frequency Switching A = 0.25 Chopping A = 0.125 very long OTF scans A = 0.5 short OTF scans A = 0.3 Using these assumptions, the formula for computing the rms noise level is for a single mixer T_rms = T_sys / sqrt(A * delta_nu * T_tot) This can be improved for all bands by a factor sqrt(2) if we assume we'll be observing with two mixers simultaneously. Numbers from the Spatial framework document (C. Kramer) ------------------------------------------------------ ---> Point source Continuum sensitivity ---------------------------------------- 450 Jy/K at 500 GHz and for bands 1 - 5 510 Jy/K at 1.9 GHz and for band 6 ---> SSB System temperatures T_sys ---------------------------------- Band 1-5 180K *(nu/500 GHz) Band 6 2000 K ---> Maximum bandwidth ---------------------- Band 1-5 2 x 4 GHz Band 6 2 x 2.4 GHz ---> Maximum Bandwidth for binning (continuum observations) ----------------------------------------------------------- 100 MHz Be careful that all numbers are SSB. Continuum sources will be seen in the two mixer sidebands. The expected continuum level in K will therefore be twice the expected SSB continuum level.