[evlatests] P-band switched power

[Rick Perley]

Despite my notes from earlier this week, P-band switched power is useful for correcting large gain changes.  Only a few of the SPWs are free of enough of RFI that the data can be suitably edited, however.  Fortunately, even a single SPW's PDif values, once cleaned up, can be used for all SPWs.

Some issues are noted here:


  1.
The most glaring is that the claimed system temperatures are outrageously incorrect:  For all antennas except a few, the claimed Tsys is a ridiculous ~10K.  The explanation is elementary — the Tcal values in the 'parmenator' have never been set to realistic values.  They are likely at the default 1K.  It would be good to get these updated.  I believe the Tcals were measured in the lab long ago.
The antennas which give at least semi-reasonable values are:  12, 23, 24, 25, and 26.  Possibly also 18 and 20, but these antennas have other problems in the PDif.  (See below).
  2.
Antennas 1, 8, and 14 showed zero-averaged noise for the PDif values.
  3.
The PDif values for antennas 18 and 20 were extremely noisy, fluctuating by factors of ~2 or more on timescales of one second.

      On a happy note, the 'PDif Compression' problem seen at all other bands is completely absent at P-band.  Observations of Cygnus A (which triple the system temperature) have no effect on the observed values of the switched power.

      Other general notes:

      1) ea02 shows gain changes of factors of 2 or more from scan to scan.  Both polarizations, and both IF sides (AC and BD) change in exactly the same way.  Gains are stable within any one scan.  The PDif values see these changes.  It's hard to understand how this can happen.
      2) ea21 has wild gain changes on the 'V' side — up to an order of magnitude.  Data look stable within any one scan.

All else looks good at P-band.



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listmgr.nrao.edu/pipermail/evlatests/attachments/20250221/6a79d57c/attachment.html>