[evlatests] Antenna Holography Review

[rperley]

I have spent the last month reviewing the four Ka-band antenna 
holography observations taken since March, 2016.  (The observations made 
in December 2014 and January 2015 are not usable, due to snow in the 
antennas.  This has a spectacular effect on the antenna patters -- see 
the upcoming memo).

During this period, six antennas have had their panel adjusted by the 
antenna crew.  In all cases but one, excellent improvement in surface 
accuracy was achieved.  I attach an illustration for ea20, showing the 
panel displacements, before and after.  The 'before' data were taken in 
mid-April, 2017, the 'after' data in mid May, 2017.  The panels were 
adjusted (in the field) on May 3 and 4, 2017.  The color wedge runs for 
-2 to + 2 mm in each.  The rms surface accuracy before the adjustment 
was 0.57mm, and 0.29mm afterwards.  (Note the small adjustment error for 
panel #16 in the outermost ring (counting clockwise) -- evidently, the 
bottom corner screw was turned too far -- perhaps it was done twice?).

ea01 was also adjusted, on 06 April, 2020.  This improved the surface, 
as expected.  However, the observation made in May 2020 was done after 
the 0.5 degree subreflector tilt, but before the 6mm horizontal shift, 
so the May images show a very large coma, which the analysis software 
was unable to adequately correct for.  No doubt, the simple Ruze 
expressions we are employing for this are not sufficiently accurate for 
large offsets in our non-axisymmetric system.  (As a reminder, the tilt 
and offset are part of an effort to understand what is wrong with that 
antenna's optics).

On the (very) bright side, all four epochs that I have reviewed again 
emphasize a key result noted earlier -- the antenna surfaces are 
extremely stable over many years.  The surface images of the antennas 
are nearly identical, for all antennas (except those adjusted), over the 
entire time period.  The stability is so good that, barring exceptional 
circumstances, there is no need to do further surface holography for the 
purposes of adjusting panels.

The six worst antennas in terms of surface rms, based on the most recent 
data, are (in order), with their current rms:

ea05    0.53mm
ea21    0.45
ea25    0.45
ea02    0.44
ea18    0.44
ea24    0.44

The five best antennas -- all of which have had their surfaces adjusted 
-- are:

ea20   0.28mm
ea15   0.27
ea26   0.27
ea17   0.24
ea04   0.24

All other antennas have surface rms values between 0.30 and 0.41 mm.

Adjustment files will be sent to Gene and Mike.  There is every reason 
to believe that all antennas can be made as good as those five in the 
list above (with the possible exception of ea01, which remains something 
of a mystery...).  The expected improvement in gain for ea05, with the 
panels adjusted, is about 25% at Q-band, and about 10% at K-band.

A memo, summarizing the entire holography effort, is being completed, 
and should be released shortly.

Rick
-------------- next part --------------
A non-text attachment was scrubbed...
Name: EA15-BEFORE-AFTER.pdf
Type: application/pdf
Size: 33876 bytes
Desc: not available
URL: <http://listmgr.nrao.edu/pipermail/evlatests/attachments/20210528/685f6edd/attachment-0001.pdf>