[daip] Re: VTESS and the primary beam

[Eric Greisen]

Frazer Owen writes:

 > 	 I am using VTESS to do an MEM deconvolution of M87.
 > I am also using the VTESS image, without convolution to selfcal the
 > uvdata. This is working pretty well; however, there is one point I
 > am worried about. The VTESS VT image is a point model of the source
 > but is it before or after correcting for the primary beam ? Since the
 > program is supposed to know about the primary beam, I assume it is
 > after this correction. However, then I need to "uncorrect" it before
 > using it as a model for selfcal. Could you verify this or correct
 > me if my assumption is wrong ?

One chases ones tail a lot in VTESS, but I think I see what is
happening.  Nothing in VTESS ever divides by a primary beam.  Instead
the VM image is multiplied by a primary beam when the residuals are
evaluated and when a convolved image is produced.  I think that it is
safe to say that the VM image contains no primary beam.  In other
words, it is not the right image to use for selfcal.

Eric