Mount design & construction by  Steve Aragon  and  Jon Thunborg;  pictures  by Kelly Gatlin

The beam mount used till now for the  25m antenna signal path was heavy; allowed  only 2 subreflector rotations to be mapped;  and required holes drilled in the feed-cone roof.  The new mount uses a central pillar with cantelever to shorten and lighten  this hardware, and allows positioning every 20deg around the feed circle.

The reference horn was on the quadrupod leg, where cables were short but the lever arm for converting pointing errors to phase offsets was 6.2m.   Now it sits  behnd the subreflector <1m off axis,  at the expense of 15m more cable.
 
 

Thumbnail.	Description
	Shot from ground, loking into dish which is tipped down.  Reference horn is 1m away from the axis,  (at top right) just outside the rotating drum that holds the subreflector - see last picture  Signal path for main antenna  (horn+mirror above feed cone) is pretty much the same as before, except for the rotatable mount - see next 2 pictures
	New rotatable mounting arrangement, note laser pointer (black finger poking up) to align the mount with antenna axis.   A similar laser mounts in the horn  throat, and reflects off a bit of (optical) mirror taped to the RF mirror to align the horn axis on the cross hairs  of the subreflector.  The typical colimation offsets for this feed are a few arcmin, not much more than those of the  regular VLBA bands.   Rotation and focus must be  re-found if the mirror is rotated, and probably should be verified each time regardless.
	Top view of mount - holes every 20deg.  The orientation (linear polarization angle) of the LNA must be rotated for each position used around the feed circle to receive  the linear polarization emitted by the satellite beacon (12.198GHz, vertical polarization at parallactic angle -10deg for the GE-4 satellite)
	Reference horn position is now at the apex behind subreflector.  To align,  put antenna at elevation=90 and use a spirit level on top face.  Note that the amplifier orientation and feed alignment need  not be adjusted once they are optimized for a given satellite.