[evlatests] K-band Sensitivity across 8 GHz

[Rick Perley]

    Using the 3C286 observation, I calibrated the noise field, and 
determined the rms per-baseline/polarization noise, for the center 
channels of each of the 64 subbands.  This used the RCP side only. 

    The attached postscript plot shows the quick results.  Key points:

    1) The overall value is about 15% higher than we expect on the basis 
of the 8-bit samplers.  Because referenced pointing was not done, nor 
did we run an 8-bit experiment at the same time, these results are not 
to be taken too seriously.  But they should be fairly close to the 
truth.  The 15% excess is the already known '3-bit penalty' from the 
samplers. 

    2) The elevated noise from 22 to 24 GHz is almost certainly due to 
ea25, which was seen to have a very noisy bandpass (this is the A2 path). 

    3) Most of the high spikes are from the DC offset problem. 

    4) The three subbands labelled 'water vapor' are noisier than I 
would have expected from last night (very dry air), but they do lie at 
the right frequency ...

    5) The lowest subbands are known to lie on a TV downlink frequency 
-- which may contribute some of the noise. 

    6) The   modest rise between 20 and 26 GHz (ignoring the 22 -- 24 
GHz piece) is expected, and is due to the decreasing aperture efficiency. 

   
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 8GHzSensK.ps
Type: application/postscript
Size: 15448 bytes
Desc: not available
URL: <http://listmgr.nrao.edu/pipermail/evlatests/attachments/20120316/d24b6962/attachment.ps>