[daip] Re: clean component tables

[Martin Shepherd]

On Thu, 18 Oct 2001, Andy Biggs wrote:
> Hello there.
> 	I've been experiencing a few problems with CC tables, specifically
> those read into AIPS from difmap. Some tasks appear to have great
> difficulty in using them e.g. FRING which complains that not all the clean
> components are the same size. Using UVSUB I was able to get it to work by
> setting cmethod='DFT'. Leaving it blank caused the same problem as in
> FRING. However, setting cmethod='DFT' in FRING did not solve the problem.

It appears that although the specification for AIPS CC tables allows
for model components that aren't delta functions, many old AIPS tasks
were only written to handle the latter. This limitation is probably
because before model fitting came along, originally there weren't any
AIPS tasks that could generate anything but delta-function components,
so nobody bothered with the more complicated case. This is thus a
limitation of some AIPS tasks, rather than a CC table incompatibility
problem between difmap and AIPS.

> 	I then wrote out the difmap CC table to file using TBOUT and
> edited this file in emacs, removing columns 4 to 7 which specify the
> ellipticity, orientation etc of the components.

This won't help you much, because you are just turning your elliptical
gaussian model components into delta functions, which probably don't
fit the data at all well.

If you have to use these AIPS tasks, the only thing that I can suggest
is that you avoid fitting anything but delta-function model components
in difmap.

Martin Shepherd   (mcs at astro.caltech.edu)