[evlatests] Gain Changes and Switched Power Effectiveness

[Rick Perley]

     Some nice news for everybody ...

     On Wednesday night we ran a 3 hour 'beam-holography' observation.  
(The goal is to accurately measure the primary beam, and near-in 
sidelobes, for proper image correction).

     The raw gain calibration -- done without applying the switched 
power -- showed some interesting rhythmic gain variations at the 
few-percent (amplitude) level.  Antennas 05 and 20 are particularly 
bad.  A plot (ea05+20-SY.png) showing the amplitude gains is attached.   
In addition to these period changes, a few other antennas showed drops 
and bumps of similar amplitude, some on very short (3 minute) timescale.

     Because beam holography requires very high amplitude stability, I 
then applied the switched power.  This requires editing these data (due 
to RFI, and other transient effects).  The AIPS program 'TYSMO' was 
fabulously effective in removing all questionable values, in all 
spectral windows (even those with known RFI!).  I'll also note here that 
all antennas had working switched power in both polarizations. No bad 
actors!  (Congratulations to the technical staff for making this work).

     The results of applying the switched power *perfectly* removed 
*all* gain variations from all antennas, all spectral windows, and both 
polarizations.  Attached is the gain plot (ea05+20+SY.png) after having 
applied the switched power.

     Remaining gain variability is at the sub-1% level.


-------------- next part --------------
A non-text attachment was scrubbed...
Name: ea05+20+SY.png
Type: image/png
Size: 76039 bytes
Desc: not available
URL: <http://listmgr.nrao.edu/pipermail/evlatests/attachments/20160115/a6fc81ed/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ea05+20-SY.png
Type: image/png
Size: 56414 bytes
Desc: not available
URL: <http://listmgr.nrao.edu/pipermail/evlatests/attachments/20160115/a6fc81ed/attachment-0003.png>