[Gb-ccb] some ccb commissioning results

[Brian Mason]

hi all- I've posted some plots of data obtained last week with the CCB on 
the GBT at

http://wiki.gb.nrao.edu/bin/view/Projects/CcbCommData

The procedure was to nod the telescope back and forth between the two 
beams, collecting data all the while. The actual antenna positions were 
used (with the known source position) to compute a template for the source 
which was fitted to the differenced data; 3C48 is used as a known source 
to scale other sources to specific intensity. Each of 16 independent 
channels are fitted independently.  (all the analysis was done in IDL 
scripts that Larry Weintraub & I developed over the past couple of weeks) 
To the "differenced beam template" we also allow a constant and gradient 
in time. Sources were selected from NVSS + OVRO 40-meter (30 GHz) observed 
sources, mostly, ones known to have fairly steep spectra and be in the 5 
to 10 mJy range at 30 GHz. We also did blank field observations in a crude 
Lead/Main/Trail fashion (the alignment of Lead and Trail is only good to 
1-3 beams in elevation which probably isn't good enough-- I need to work 
on this but it's a bit tricky with our system & method of observing). 
Another item for upcoming tests is we really should be using a symmetric 
NOD pattern (we found that even in 80 seconds under excellent photometric 
conditions the gradient in the difference data is quite significant 
statistically, and biases our results if not included in the fit).

In the data plots white is the data & green is the fit; subpanels are 
individual CCB input ports (RF detectors).  At the bottom of the page I've 
collated results on individual sources vs frequency. Here different colors 
are independent observations of a source. Recall that for a given 
observation (color) there are four independent RF detectors that nominally 
see the same frequency range-- these are spaced out in frequency for 
clarity.

We are still processing the results but a couple of conclusions thus far:

*the formal error in the fitted flux density from a 2 minute NOD 
observation, fully calibrated, for one *single* CCB port, is about 0.2 mJy 
(up to 0.4 for high frequency channel). This is close to the expected 
theoretical noise.

*It's not clear that our actual constraint on celestial flux density is 
quite that good but it's not much worse. See for instance the blank field 
observations, channels > 28 GHz (however see next point)

*The consistency of the low frequency channel is poor, both on blank sky 
and on-source.

We have a full night (possibly more) later this week as well on which 
we'll do more of the same.

  cheers,
   Brian

--------------------------------------------------------------------
Brian Mason                           | office: +1(304)456-2338
Associate Scientist                   | fax:    +1(304)456-2229
National Radio Astronomy Observatory  | mail:   PO Box 2 
bmason at gb.nrao.edu                    |         Green Bank, WV 24944
http://www.gb.nrao.edu/~bmason/       |
--------------------------------------------------------------------