[daip] PCAL and models

[Bill Cotton]

Martin,

   I'm not sure how difficult it is to "unfix" PCAL but an
alternative, which I think still works is to use UVSUB and subtract
the Q and U clean models from the UV data and then force PCAL to not
fit for the polarization.

-Bill

Martin Hardcastle writes:
 > On Fri, Dec 16, 2005 at 09:09:58AM -0500, Bill Cotton wrote:
 > >    This used to work but it's been "fixed" since then.
 > 
 > Is it possible to un-fix it?
 > 
 > I have poked around a bit more in the code (thanks for the pointer,
 > Eric) and yes, it's true that XSMOD is used -- it is put into VFLUX
 > just as for the ORI- case, only within PCLFLX. It is then passed to
 > LPCALC (as VST...) but LPCALC doesn't make any use of anything other than
 > VFLUX(1,...) i.e. STOKES I. It looks to me as though LCALC should be
 > being passed a model derived from XSMOD in VSTMOD, not VST, but I
 > could be completely wrong.
 > 
 > In any case, if it's broken and there's no prospect of a fix, can
 > the documentation be changed to reflect that?
 > 
 > Thanks
 > 
 > Martin