[daip] aips fring

[Eric Greisen]

On 06/30/2016 09:44 AM, cao hongmin wrote:
> Hi Eric,
>
>
> I have a minor question needing your help.
>
> When we run fring without using a calibrator model, a point
>
> model is used and it is put at a position which has the same
>
> coordinates as the calibrator.
>
> (1) The problem is what will happen if the coordinators of the
> calibrator is not accurate (?)
>
> e.g. the actual position of the calibrator is hundreds of
> milli-arcseconds away from its prior
>
> coordinates.
>
> How far the deviation will cause an obvious difference compared to we
> have accurate coordinates (i.e. at a level of mas) (?)
>
> (2) We could use a calibrator model to improve the fringe-fitting results.
>
> But, if the coordinates are not accurate, whether we still can use a
> model to re-run task fring (?)

Not so minor actually.  100's of milli arcsec will introduce a delay 
error and rate that will badly affect the results from fring if you wish 
to apply those results to other sources.  Using a model with the 
components properly offset (thus at the correct actual position) will 
work.  Otherwise, this does not seem like a good choice of calibrator.
Perhaps Amy will have further thoughts (she is included in the daip copy).

Eric Greisen