[evlatests] New 'Spectral Sweep' Results

[Rick Perley]

    Ken ran the 'spectral sweep' program.  This steps the array through 
the entire frequency range -- from 1 to 50 GHz, in 1 GHz steps.  The 
spectral resolution is 125 kHz.  Fringe tracking is turned off (by 
telling the system that we're pointing at the north pole). 

    I generated two sets of plots, showing the results.  These show an 
arithmetic (scalar) sum over all baselines.  No attempt was made to 
either edit or calibrate the data.  The results are shown separately for 
RCP and LCP (although the difference, as expected is trivial -- the RFI 
comes in through distant sidelobes, which are randomly polarized).   The 
vertical axis is in logarithmic units -- each unit increment is a factor 
of ten higher spectral power.  The noise floor roughly represents the 
effective antenna SEFD -- ~300 K at the lower frequencies. 

    The results can be found in:

    /DATA/SECHELT_1/FITS/RFI-18JAN13-1GHZ.eps

    There are 98 plots in this file, arranged in RR,LL pairs.  Each 
shows a single 1.024 GHz-wide chunk of the spectrum. 

    I used 'convert' to convert this file to pdf -- the name is the same 
with the .pdf extension in place of .eps.  However, to 'acroread', the 
plots (except the first) are truncated on the right and bottom sides.

    I also converted the plots to .png format.  These are in the same 
directory, with the same name, but are now separated into 98 individual 
plots, named 'RFI-18JAN13-1GHZ.nn.png'.  The RCP plots have nn even, the 
LCP are nn odd.  The lower frequency edge for the RCP plots is given by 
nn/2+1.  The lower frequency edge of the LCP plots is given by 
(nn-1)/2+1.  This format seems to have translated correctly.