[evlatests] A new kind of 'wobble'?

[Rick Perley]

    The database taken by Michael last week (8 subbands of 128 MHz 
each,  512 channels/subband, RCP only, 1 sec. integration, at L-band) 
has revealed a heretofore undetected phenomenon -- wobbles in both time 
and frequency on a single baseline. 
   
    The AIPS program SPFLG displays a 'waterfall' type plot, showing the 
individual baselines (amplitude or phase) as a greyscale 2-D plot, with 
channel (frequency) on one axis, and time on the other.  I was viewing 
some of the data, idly stepping through the 66 individual spectra, when 
one stood out on its own:  Baseline 5 x 15 (W8 x W6) has a remarkable 
fine-scale standing wave, of frequency periodicity ~1.4 MHz, and very 
high amplitude in both visibility amplitude and phase.  (Amplitudes are 
+/- 50% about the mean, phases are +/- 25 degrees about the mean).  
These 'fringes' are seen in all eight subbands with the same frequency 
spacing. 

    The positions ('phase') of these fringes, however, is *not* constant 
in time, but drifts slowly and steadily.  (So, in a 'waterfall plot", 
the corrugations are tilted away from the vertical by a small angle).   
Curiously, the temporal drift rate is *not* the same in each subband:  
As a function of time, the periodicity in subband 2 (1180 MHz) is 1.67 
minutes, while in subband 8 (1950 Mhz), it is 1.0 minutes --  to the 
accuracy of my estimations, the time periodicity is proportional to the 
observing frequency. 

    Only this baseline shows this effect.  All others are stable, and 
any 'periodicities' seen are clearly due to antenna bandpasses. 

    I attach a frequency plot, showing the phenomenon in a single point 
in time. 


-------------- next part --------------
A non-text attachment was scrubbed...
Name: WIGGLE
Type: application/postscript
Size: 62971 bytes
Desc: not available
URL: <http://listmgr.nrao.edu/pipermail/evlatests/attachments/20091112/02ef55dc/attachment.ai>