[evlatests] summary

[Ken Sowinski]

Barry has explained that there is another MIB timekeeping error
plagueing us.  Until a solution is available I have turned off
phase switching so that the antennas will at least appear to work
and routine things can still be carried out.  Now that it is
clear that we can observe reliably with phase switching disabled
if need be, I am happy to withdraw the objection I offered to moving 
the antennas.

Pathologies.
13A and C are dead at L band.  No signal at the sampler or the T5.
14D is very bad at X band.
14C is just bad at X band.  It is not clear if this is a receiver
problem or something downstream.
16B is bad at C band.  Likely because the sampler is overdriven.

To keep Rick happy relative G/T was measured with ANTSOL while 
observing 3C84 near transit.  All comparisons below are measured
in power and ignoring the problems reported above.
C band.  13 and 14 are the best.  Most VLA antennas are 10-20 
percent worse.  16 is poor but better than most VLA antennas.
X band.  13 is among the best.  16 is fair and 14 is poor.
L band.  All antennas are about 75 per cent of the best VLA
antennas.
Reliable measurements at K and Q band await better pointing.
However 14 is not bad, even at Q band, but 16 needs better pointing
to be useful.


when the system was first brought up this morning 16 was not 
available because its D rack was in use for PT tests, 13 worked
and 14 did not.  The problem with 14 was that the FRM was in the 
wrong place.  The MIB had gotten commands, thought it had passed
them along to the FRM, but the FRM error signals were large for
both focus and rotation.  The script was started again and all 
worked as expected.  My naive suspicion is that there is a problem
in the FRM controller.  This has been seen on ocassion for years with
VLA antennas; I have always blamed it on lost commands to the antenna,
but now that we see what appears to be the same thing in EVLA antennas 
suspicion starts to fall on the FRM.  I wonder if there is a small 
window in time when it will accept a command but not use the information 
to drive the hardware.