[evlatests] summary, Feb 6 2007

[Ken Sowinski]

This message covers several topics.

Software
Some time was devoted to installing and testing the next
version of the executor and debugging the spectral line
delay solver in 4 IF mode.  While doing this I was able
to compile a list of the the apparant delay error between
VLA and EVLA antennas as the BW is narrowed from 50 MHz
where, by definition, the difference 0.0.
BW   Apparant delay difference (refant = EA24)
==   ========
50 MHz  0 ns
25      2.5
12.5   -5
6.25  -35
3.1  -110
1.5  -270
0.8  -470


Antenna 17, K band
Antenna 17 now works at K band.  Delays and focus are set;
pointing is good enough that it should work when we next
do collimations.


Antenna 17, X band
X band is very weak, down by an order of magnitude in G/T.
There is too much power in the crossed hands, 10 to 20 per cent,
and T_sys looks higher than it ought to be.  It certainly looks
like there is a something wrong the the horn.


Antenna 26
The problem with IF B has gone away and it looks fine now.
There is, however, a problem with IF C.  At X band the
correlated amplitude and the SD voltage are both very
unstable on a time scale of seconds.  It seems there is
something besides the expected noise getting to the T5.
My default explanation is that this is related to the test
setup for the phase stability tests, but it might be something
real.


Relative G/T performance
Not too many surprises here.
C band.  13, 14, 16, 18 all better than VLA antennas, but 24 is
only about the same as a VLA antenna.

K band.  Pointing was not done, but EVLA antennas were not much
different from VLA antennas.

L band.  Mediocre, as expected.  26 did not work at L band.

X band.  17 and 26 were discussed above.  In addition the LCP
side of 14 is down by about a half in power compared the the
RCP side and all the other antennas.

Q band.  18AC did not fringe.  I was able to get it to work by
tuning the L302 away by a GHz or so.  All EVLA antennas worked
but 13, 14 and 18 were weak; again probably because pointing
was not done.  16 and 24 looked quite good, even without pointing.