[evlatests] antenna 28 pointing still bad

[Ken Sowinski]

On Tue, 11 Nov 2008, Ken Sowinski wrote:

> Meanwhile I can say two things about how the antenna is
> misbehaving.  These signatures amy be a consequence of the
> azimuth encoder tracking error, or be independent problems;
> it is hard to distinguish.  1) The measured azimuth pointing
> is proportional to elevation and varies about an arc-min
> between elevations of 15d and 115d (over-the-top).  It is
> *not* fit by the usual perpendicularity term.  2)  The
> measured elevation pointing error is bistable; between about
> 70d and 150d it is close to zero, elsewhere it is about
> -0.5 arc-min.  Neither of these two signatures can be fit
> by any terms in the standard pointing model.  I have plots
> illustrating these if anyone wants to look at them on
> Wednesday.

The above statement was based on a five hour pointing run several
nights ago.  Since then the az eoncoder has been replaced. 
A three hour pointing run last night shows the antenna to be no
better and the two symptoms described above are still there.

A more careful examination of the data taken for the pointing
runs suggests that the antenna misbehaves at low elevations.
There were many, but not all, scans below 20 degrees in which
the amplitude response for antenna 28 was essentially zero, or
more than an order of magnitude smaller than the other antennas.
There is weak evidence that the degradation is not caused by
pointing.  For one scan there was sufficient response on one
baseline to see that the pointing offsets were not large at all.