[evlatests] The Return of integer zeros ...

[Bryan Butler]

i had a test from last friday that had very odd behavior in the BD pair. 
  here are my notes from the data reduction yesterday:

    very weird effect in the 2nd baseband.  for the LL correlations
    (IF D), there are a bunch of missing baselines (missing in
    the sense that they are blanked) for the 2nd half of the SB, which
    was after a reference pointing scan.  for the RR correlations,
    (IF B), this is true throughout the SB.

this was Ku-band, with AC tuned from 13-14 GHz, and BD tuned from 
16.5-17.5 GHz.

so i just had a look at the BDF2AIPS log file:

Wrote 963600 visibilities, drop 50574 all zero.

so more than 5% identical 0's.

i can make a list of the baselines that have the 0's if it will help.

	-bryan


Rick Perley wrote, On 11/7/12 7:55 :
>      We ran a 'PDif Compression' test last night.  In reviewing the data,
> I find that large fractions of some of the scans contain 'integer zeros'
> (and hence are removed by BDF2AIPS).  In column listings, some baselines
> are fully present, but about half the baselines have only ~25% of the
> data.
>
>      The data were taken in a simple mode:  Only two subband pairs, one
> from the AC IFs, the other from the BD IFs.  The problem is not
> uniformly seen across all bands --
>
>      Only for K, Ka, and Q bands, and only on IF 'D'.
>
>
> _______________________________________________
> evlatests mailing list
> evlatests at listmgr.cv.nrao.edu
> http://listmgr.cv.nrao.edu/mailman/listinfo/evlatests
>