[evlatests] Basic results from first Orion database

[Rick Perley]

    Michael has observed the Orion core in three 'chunks', each 1.024 
GHz wide.  I have done the basic calibration for the first of these -- 
23630 to 25655 MHz. 

    All antennas fringe nicely, although antenna 27 is (as seems to be 
usual, for reasons unknown) weak by a factor of at least 10. 
    There are a very small number of 'zero records' -- 0.018% of the 
visibility data are zeroed.  These are easily identified and removed.  I 
see no evidence for any significant number of reduced visibilities that 
cannot be ascribed to antennas off source, or other antenna-based effects. 

   I used POSSM to check a few of the raw spectra, looking for 
'displaced' subbands.  All spectra were continuous in amplitude and 
phase across subbands. 

    Basic calibration was done:  FRING to find the delays.  SETJY/GETJY 
to establish the flux densities.  CALIB to calibrate the subband gains, 
BPASS to remove the bandpass shapes. 

    The calibrated spectra are -- with a few exceptions -- smooth in 
amplitude and phase across subband boundaries.  The only exceptions are 
between subband 1 and 2, where some baselines show a small apparent step 
in amplitude.    No explanation is offered, other than the obvious one 
that this was introduced by the calibration, since no step occurred in 
the raw data.  Subband 1 is peculiar  -- there is clearly very high 
aliasing of external signal on its lower end -- perhaps some other 
effects are making antenna-based calibration ineffective in this 
subband.  (That is, the data don't close). 

    At least 9 spectral lines are visible in short (couple of minute) 
vector integrations. 

    I tried making an image of the continuum and of one of the lines -- 
but the results of this were puzzling.  It would be good if somebody who 
has knowledge of the expected structure could get involved now.  

    Rick