[evlatests] Correlator Offsets, (cont.)

[Rick Perley]

    A test was run this morning, to see if the infamous offsets remain, 
following some work yesterday by Ken and Ray. 

    Ninety minutes was spent with the correlator, at 25 MHz BW, and mode 
'4'.  All four IFs shows offsets, in the form of 'hash' at the center of 
the field, of maximum amplitude a few hundred microJy.  The true noise 
in the field is about a factor of 10 below this.  I made no attempt to 
deconvolve (the effect is perfectly clear without having to do this) -- 
a number of background objects are clearly visible.  The hash is present 
with and without calibration.
    Ninety more  minutes were spend, in the same mode, with the 
correlator self-test turned off.  These images showed no hash at all.  
Turning off the self-test certainly eliminates the problem.

    Ken has inquired as to how positive we are that the effect is absent 
when observing with 12.5 MHz BW.  To confirm this, I remade the image 
from my test last month -- 4.5 hours on a blank field.  There is 
absolutely no sign of the offset 'hash'.

    Rick