[evlatests] P-Band Beam Cuts -- a major surprise!

[Rick Perley]

    All:

    I'm sending this to a wider audience than usual, since the results 
below are I think of general interest, and could benefit with a wider 
group. 

    I finally got around to doing basic holographic beam cuts through 
the principal planes of the VLA antennas at P-band.  The results are 
quite surprising. 

    I made two cuts, each 22.8 degrees in length (11.4 degrees on each 
side of the beam axis), one in the horizontal, the other in the vertical 
plane.  There are 20 'stopping' places, each 10 seconds long, on each 
side of the beam center.  Hence, each of these is another 0.57 degrees 
further along.  Sampling was done at 1 Hz. 

    The standard editing and calibration were done.  The target source 
was 3C147. 

    Attached are four .eps plots: 

    1) PolR-El.eps ('R' polarization, in the elevation plane)
    2) PolL-El.eps  ('L' polarization, in the elevation plane)
    3) PolR-Az.eps ('R' polarization, in the horizontal plane)
    4) PolL-Az.eps ('L' polarization in the horizontal plane). 

    I *think* that 'R' corresponds to horizontal polarization, and 'L' 
corresponds to vertical.  (I would be good to finally settle this issue). 

    These plots are in *voltage*,  and are normalized to the beam 
center.  So 0.1 on the vertical axis corresponds to -20 dB in power, and 
0.3 is -10 dB.  The half power is thus at 0.71 on the vertical axis.  
The horizontal axis is in time units -- 30 seconds  is 1.71 degrees. 

    The key points to note are:

    All cuts show the expected plateau, which I believe is the primary 
result of the defocussed feed (the bottom of the subreflector is about 
60 cm below the prime focus).   But what I hadn't expected was the very 
large asymmetry in the two planes:

    A) For the horizontal cuts, the 'R' polarization (horizontal?) has a 
smooth plateau with a peak at about -14 dB below the peak.   The 'L' 
polarization has a very much higher plateau -- down only about -6 dB! 
    B) For the vertical cuts, the situation is exactly reversed -- the 
'R' polarization plateau is now higher,  -9 dB, while the 'L' 
polarization plateau is reduced to about -15 dB. 

    It is possible we have elliptically shaped plateaus -- or possibly 
'islands' which extend along these principle planes. 
    We haven't noted these curious effects in the past, since we used 
RCP and LCP, which combines the two linears.  (The asymmetries would 
show up in the cross-polarization, but little atttention was ever paid 
to these channels back then).

    These results are not specific to the pair of antennas chosen for 
these plots.  All antennas, and all subbands give the same results. 

    In talking to Ken, Barry, and Vivek, various opinions as to the 
origin of this asymmetry were made: 

    1) The quadrupod legs.  A very likely contributor.
    2) The asymmetric subreflector.  It is currently so the asymmetry is 
in the vertical plane.  An easy test is to rotate it by 90 degrees. 
    3) The guy rods -- we think these are only in the vertical plane -- 
their effect should be to 'widen' the effective width of the vertical 
quad legs. 

    It would be very useful to undertake a proper diffraction 
calculation of the beam patterns.  We (or at least I) have very little 
idea of what we should expect to see.  \

    After I get back from the AAS meeting, I'd like to do a full 41 x 41 
dish holography scan.  This will show us, with great precision the 
actual illumination patterns for each polarization. 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PolL-El.eps
Type: application/postscript
Size: 31368 bytes
Desc: not available
URL: <http://listmgr.nrao.edu/pipermail/evlatests/attachments/20131231/9dc20ae7/attachment.eps>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PolL-Az.eps
Type: application/postscript
Size: 28029 bytes
Desc: not available
URL: <http://listmgr.nrao.edu/pipermail/evlatests/attachments/20131231/9dc20ae7/attachment-0001.eps>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PolR-El.eps
Type: application/postscript
Size: 31478 bytes
Desc: not available
URL: <http://listmgr.nrao.edu/pipermail/evlatests/attachments/20131231/9dc20ae7/attachment-0002.eps>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PolR-Az.eps
Type: application/postscript
Size: 28029 bytes
Desc: not available
URL: <http://listmgr.nrao.edu/pipermail/evlatests/attachments/20131231/9dc20ae7/attachment-0003.eps>