[evlatests] An odd (new) ACU issue

[Rick Perley]

     I took some S-band holography data last week -- the goal is to 
better understand our beam polarization characteristics.

     The holography observations used a 2.5 arcminute stepsize, with the 
fast rastering done in the vertical direction.    The vertical raster 
direction was chosen to (more or less) remove the large azimuth motions 
that come with holography grids taken at high elevation.  The sources 
chosen to grid were 3C147 and 3C138 -- the elevation range of the 
observations spanned 47 through 74 degrees.  Each raster point was 
observed for 10 seconds.  The grid was 11 x 11.

     In reviewing the 'strip scans', a curious dependency on elevation 
was found for the 'new ACU' antennas.  I attach four plots to illustrate 
the point.

     The first plot (OLD24-El74.png) shows the typical profile for the 
central vertical cut for an 'old ACU' antenna (ea24).  Each sample 
duration is 1 second.  There is a sharp overshoot, lasting three to four 
seconds, after which the amplitudes are stable. This profile is 
completely elevation independent -- all plots for all 'old' antennas at 
all elevations look the same.

     The behavior for the 'new-ACU' antennas is quite different.

     The second plot (NEW21-El47.png) shows the profile at an elevation 
of 47 degrees.  It shows much better behavior than the 'old' antennas - 
as expected.  All the 'new' ACU antennas behaved like this at this 
elevation.

     But I was quite surprised to see that at high elevation, the 
behavior is quite different.  The third plot (NEW21-El74.png) shows 
there is a large, slow, and seemingly undamped oscillation of period ~10 
seconds.  All the 'new' antennas which were not reference antennas in 
this experiment showed this effect (these were 21,1,8 28 and 17), 
although ea21 was by far the most pronounced.  To illustrate, the fourth 
plot, for antenna 17 and elevation 74 degrees (NEW17-El74.png) is shown.

     The question to me is:  why do the new ACU antennas have an 
elevation dependence in their response for these elevation steps? A 
notable change in behavior is to be expected for azimuth offsets, as the 
'stepsize' is a strong function of elevation.



-------------- next part --------------
A non-text attachment was scrubbed...
Name: OLD24-El74.png
Type: image/png
Size: 17644 bytes
Desc: not available
URL: <http://listmgr.nrao.edu/pipermail/evlatests/attachments/20171211/0016383c/attachment-0004.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: NEW21-El47.png
Type: image/png
Size: 16635 bytes
Desc: not available
URL: <http://listmgr.nrao.edu/pipermail/evlatests/attachments/20171211/0016383c/attachment-0005.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: NEW21-El74.png
Type: image/png
Size: 17548 bytes
Desc: not available
URL: <http://listmgr.nrao.edu/pipermail/evlatests/attachments/20171211/0016383c/attachment-0006.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: NEW17-El74.png
Type: image/png
Size: 17103 bytes
Desc: not available
URL: <http://listmgr.nrao.edu/pipermail/evlatests/attachments/20171211/0016383c/attachment-0007.png>