[evlatests] [Fwd: survey of l302 glitches]

[Rick Perley]

    I replicate Ken's message, with comments on the behavior of antennas
13 and 18 following each of the actions listed by Ken. 

-------- Original Message --------
Subject: 	[evlatests] survey of l302 glitches
Date: 	Mon, 3 Jul 2006 10:50:03 -0600 (MDT)
From: 	Ken Sowinski <ksowinsk at nrao.edu>
To: 	evlatests at nrao.edu



Rick and I used an hour of dynamic time this morning to try
to understand how the amplitude glitches interact with phase
switching.  All antennas except 24, which unfortunately was
locked out, were avaialable.  Rick prepared a file which
was simply successive 30 minute scans looking at 3C84.  I
report here on the sequence of events during observing and
will leave the detailed reporting of behavior to Rick who had
a more synoptic view than I.

However my impression there has been more than one problem.
Phase switching and fringe rotation commands seem to interact.
13 can be fixed by disabling phase switching.
18 can be altered, but not fixed by disabling phase switching.


IAT   Test response to phase switching
---   --------------------------------
1530  antennas were on source

	During this interval, both 13 (all IFs) and 18 (BD only) 
had regular drops.  
	Antenna 13's drops were for a single 0.4 second record, 
and occurred on x0.4 seconds (i.e., on 00.4, 10.4, 20.4, etc.) 
seconds.  All drops (as seen on VLA-coupled baselines) were by
about 50% in amplitude).  
	Antenna 18's drops were 'double', with a deeper drop 
at x0.4 seconds, followed by a shallower drop at x0.8 seconds.  
Deeper means the amplitude dropped to 40% of normal amplitude, 
shallower means the amplitude dropped to about 55% of normal 
amplitude.  

1536  disable phase switching for 13BD at the L302

	During this interval, 13BD's fringes went away.  
13 AC continued with same behavior as listed above. 
	There was no change in antenna 18's behavior.

1539  disable phase switching for 13 at the correlator

	Antenna 13, IFs B and D, returned *with no drops*.
Antenna 13, IFs A and C, remained dead.   

1545  disable phase switching for 18BD at the L302

	Antenna 18's fringes on BD went away after this action.  
Antenna 13's behavior remained unchanged (AC gave no fringes,
BD gave drop-free fringes).  18 AC still gave good fringes.  

1547  disable phase switching for 18 at the correlator

	Antenna 18's fringes on BD returned -- *but with the same
drop behavior as noted above.  Antenna 18 AC fringes disappeared.

1549  disable phase swithcing for 13AC and 18AC at the L302

	Antenna 13 AC fringes returned -- without drops.  
	Antenna 18 AC fringegs also returned, without drops (but 
none have ever been seen on 18 AC).  

	Something happened at 15:56 -- not recorded by Ken -- 
which changed the nature of the drops on antenna 18, BD.  
Instead of all drops being in pairs, most are now single and 
quite weak -- only a few percent.  Approximately every fifth drop
is double with the same depths are recorded earlier.  

1600  New scan resets the EVLA
1601  enable phase switching for 13 and 18 at the correlator

	Both antennas drop, but with somewhat different characteristics 
to those at the beginning of the experiment:

	Antenna 13 drops at x1.3 seconds, to an amplitude 0.76 of the
normal (to VLA antennas).  All four IFs behave the same way.  

	Antenna 18 (BD only) drops four times at x1.3 seconds to an
amplitude of 0.62 of normal, then comes a double-drop, at x1.3 to 0.26
of normal amplitude, and x1.7 to 0.62 of normal.  

       Test response to fringe commands
       --------------------------------
1603  Operator interrupts script for EVLA antennas

	There are *no drops* on either antenna 13 or 18 during
this interval.  

1606  I start a private executor running the same script

	There are no drops on either antenna during this period.
	However -- it is clear that antenna 18 is not tracking
during this period, as the fring amplitude slowly, but increasingly,
decline during this period.  
  
1612  I kill the private executor


1613  Operator restarts the EVLA script

	Drops apppear again -- but not they are at x2.9 seconds  
(02.9, 12.9, 22.9, etc.).  Antenna 18's drops are now all 'single', 
but have two different amplitudes, with 90% of the drops being deeper
(to 65% of normal), and 10% being shallow (down by 10%).  

       Phase switching exercise again
1615  disable phase switching for 13AC at the L302

	13 AC fringes vanish. 

1616  disable phase switching for 13 at the correlator

	13 BD fringes vanish also, at 16:16:20.   Antenna 18
fringes vanish -- all four IFs.  

1617  disable phase switching for 13BD at the L302

	13 AC fringes reappear, with no drops.  13BD fringes are
absent.  Antenna 18 fringes reappear, with same characteristics 
as before.  (Drops on BD of two different amplitudes, no drops on AC).

1618  disable phase switching for 18BD at the L302

	Antenna 13 has clean, drop-free fringes on all IFs.  
	Antenna 18 BD fringes vanish.  

1620  disable phase swithcing for 18 at the correlator

	Antenna 18 AC fringes disappear.
	Antenna 18 BD fringes reappear, but with different nature.  
The Deep Drops are gone, leaving only the shallow (~10%) drops.  

1622  disable phase switching for 18AC at the L302

	Antenna 18 AC fringes reappear.
	Antnena 18 BD fringes remain as before (shallow drops, every
10 seconds).  


1631  new scan resets everything
	
	And all drops reappear, this time at x1.2 seconds.  
	Antenna 13's drops are all of the same depth (by 0.75 of normal 
amplitude to VLA baselines).
	Antenna 18's drops are of two different heights.  The more shallow
appear at every 10-second interval:  0.2, 10.2, 20.2, etc. -- *except*
when there is a double-drop.  In this case, the stronger drop (to 0.25 of
normal amplitude) occurs at x0.2, and the shallow drop follows immediately
at x0.7.  About 1/3 of the drops are 'double', but there is no clear
pattern of singles or doubles.  

	Overall, it seems clear that antenna 13's drops are of a single
nature.  They are always the same on the four IFs, and always of the
same depth.  Antenna 18's drops are more complicated, and may have more
than one origin.


_______________________________________________
evlatests mailing list
evlatests at listmgr.cv.nrao.edu
http://listmgr.cv.nrao.edu/mailman/listinfo/evlatests