[evlatests] Bizarre behavior on 10jun08

[Michael Rupen]

Hello pundits,

   I've been looking at yesterday's [10jan08] AR642 run (continuum, X & K bands,
3.3s integrations, mostly fast-switching, 16:52-17:52 IAT), and have
run into two bizarre problems.

1- The u, v, & w are set to 0.0 for all baselines, for 10 seconds in the
   middle of several (both fs and non-fast-switching) calibtrator scans.
   At the same time:
   - the amplitude goes way down (factor 10-40 on the couple baselines I
     checked in detail)
   - the phase jumps (always) and starts to wander much more than normal
     (sometimes).
   Then everything resets: uvw, amp, and phase return to their previous values.

   I've attached a plot of u vs. time for a couple examples to illustrate
   this effect.  This shows a fast switching scan from 17:25 to 17:35 (note
   the u->0 in the last subscan), followed by two scans on nearby calibrators
   (note the u->0 in the first of these scans).  There's another u->0 after
   this, on 1331+305, after this window.

   This occurs at both X and K bands.

   I noticed that I received no pointing solutions for one scan, while
   two other pointing scans worked fine.  I assume this is another symptom
   of the same problem.

2- On the first (fast switching) scan, there is a labelling problem.  Let's
   say I'm fast-switching between A and B.

   What really happens:
         AAAAAA A-B   A-B   BBBBBBBB B-A    A-B   AAAAA ....

   How this is labelled:
         AAAAAA  B   blank  BBBBBBBB  A   blanked AAAAA ....

   So the first "source change" integration is mis-labelled as belonging to
   the succeeding source; while the second "source change" integration is
   correctly flagged.

   The second and subsequent scans show the same effect, but most (not quite
   all!) of those data are (correctly) flagged on this first "source change"
   integration.

Onwards...

           Michael
-------------- next part --------------
A non-text attachment was scrubbed...
Name: tmp.ps
Type: application/postscript
Size: 125583 bytes
Desc: ar642_10jan08_u0.ps
URL: <http://listmgr.nrao.edu/pipermail/evlatests/attachments/20080111/a9db1892/attachment.ps>