Tsys measurements for VLBI at Arecibo

To measure Tsys, we record simultaneously and continuously two-polarization continuum data along with all VLBI observations at Arecibo. This is achieved through square-law detectors, and
analog-to-digital converters.

Here is an example:

For a particular C-band VLBI, the band centre of the receiver was set to 5050 MHz, which was down converted in the Gregorian dome, and then brought down to the control room. There, it was branched out to two parallel stages:

(1) at 750 MHz to feed the VLBA4 DAR,

(2) down conversion on a parallel track to 260 MHz for local use,

We passed this path (2) 260-MHz IF through a 40/5-MHz filter, amplified it, and fed the signals for the two polarizations to square-law detectors, followed by integration via a 200-msec time constants. The two polarizations were then recorded to disc separately through the analog-to-digital converters recording a data point every  100 msec.

So, the Tsys measurements are made only over over a bandwidth of 40/5 MHz at a frequency set by ourselves (we try to fix it at the center of the entire band used in a particular VLBI observation).

Using an off-source section of the continuum data at a known Az and Za, and the "standard Tsys and Gain curves" the raw-counts are converted to SEFD values. Also, as the circular polarization is created out of two native linears, we only provide an averaged SEFD which applies to both polarization channels.

So, the values you receive are ON-SOURCE (continuous) SEFD's. In ANTAB, this translates to using the Keyword, TIMEOFF = 0.0.
Here is the preamble of a typical "dearpi" letter that explains the facts described above in more detail:
!--- I M P O R T A N T --- please read the text before using the table--

! Dear P.I.
! These are actually SEFD values. With Gain=1, these could be used
! as Tsys for VLBI calibration in AIPS. As Arecibo Telescope's gain is both
! Az and Za dependent, and AIPS can not handle such a function, we
! have decided to provide the users with the SEFD values.
! Please Do not use "2pt" option, USE "self" in CLCAL
TSYS AR FT=1.0 INDEX = 'R1:4' INDEX2 = 'L1:4'/
! This way, one SEFD measurement is read into 4 RCP and 4LCP IFs.
! If you have more/less IFs(= n), then change 1:4 to 1:n in the line begining
! with TSYS above.
! day UT Tsys-av(=SEFD)
244 4:03:10.00 6.83870
244 4:03:20.00 6.84017
244 4:03:30.00 6.84219
244 4:03:40.00 6.84431