Date: Tue, 12 Oct 2010 19:27:57 -0400 (EDT)
From: Jun-Hui Zhao 
To: Bob Sault 
Cc: melvyn wright ,
     Dan Marrone , smamiriad@cfa.harvard.edu,
     Peter Teuben 
Subject: Re: miriad beta trouble

Hi Bob,
     Thanks for the email. We will look into the details.
     Jun-Hui


On Tue, 12 Oct 2010, Bob Sault wrote:

> Jun-Hui, Mel,
>
> For the SMA's nasmyth mount, the rotation of the feeds with respect to the
sky
> happens in two stages. At both stages, polarimetric leakage can
be introduced.
> Because these two rotation points change independently, in principle you need
to
> account for them separately in the polarimetric calibration. There are two
> rotation angles.
>
>
> For the SMA software I developed in late 2006, the uvvariable "chi" was (as
> always) the total rotation of feed with sky, and "chi2" was the angle at the
> second rotation point. That is, the angle at the first rotation point is
> chi1=chi-chi2.
>
> I think it is better to have a new uv variable to encode the second angle,
> rather than to have special code to compute it. The reasoning is much the
same
> as the reason that we have "chi" rather than to recompute it all the same: on
> the one hand it is inconvenient and the other hand it is better to abstract
that
> sort of telescope-specific detail to somewhere else.
>
> Dan, I think, had quite limited success on real data in using the code that
> accounted for the two rotation points. So it is all a moot issue to some
extent.
>
> In addition to handling the two rotation point calibration, the software I
wrote
> for the SMA polarimetry rather neatly handled time-shared feed systems (ie
where
> a signal path for a particular antenna is switched between different
> polarizations with time). This software relaxed GPCAL's insistence that all 4
> polarization products had to be present at any given time, and also made
> (approximate) conversion to Stokes parameters rather seamless.
>
> Incidently, I do not recall whether the nasmyth mount in the old OVRO
antennas
> has one or two rotation points. I have a vague recollection that the receiver
> package is fixed to (ie rotates with) the elevation axis - which would mean
that
> there is only one rotation point? 
>
> Best regards
> Bob
>  --
> -------------------------------------------------------
>
>
> Phone: +61-2-98721028
> Email: rsault@nrao.edu
> Web: http://astro.ph.unimelb.edu.au/~rsault
>
>
>
> ----- Original Message ----
> From: Jun-Hui Zhao 
> To: melvyn wright 
> Cc: Dan Marrone ; smamiriad@cfa.harvard.edu;
> BobSault ; Peter Teuben 
> Sent: Wed, 13 October, 2010 9:09:21 AM
> Subject: Re: miriad beta trouble
>
> Hi Mel,
>     Good suggestion! We should keep one variable so long as all the
> parties agree on it.
>     Jun-Hui
> > >