Subject: pattern-mismatch discovered on 990419s135
        Date: Sun, 25 Mar 2001 17:16:43 -0800
        From: Howard McCallon <hlm@ipac.caltech.edu>
Organization: Caltech/IPAC
          To: roc@ipac.caltech.edu, beck@ipac.caltech.edu
          CC: hlm@ipac.caltech.edu, davy@ipac.caltech.edu,
              hurt@ipac.caltech.edu, slw@ipac.caltech.edu,
              gene@ipac.caltech.edu
 

In the course of preparing input files for the Martinizing effort, I came
across a scan with very bad position reconstruction.  This scan (990419s135) is
a rare pattern mismatch, which slipped through.  The positions are normal at the
south end of the scan but may be as bad as 12 asec (or more) off at the north
end.  This was spotted because of missing overlaps for 3/4 of the scan.  I've
checked for other scans with the same characteristics and this appears to be the
only one.

I was able to successfully reconstruct this scan in stand-alone by making a
minor change to the pass1 PosFrm namelist.  For the pipeline rerun I suggest
adding the following two lines to the pass1 namelist just before the end
statement

       ,tcfrac = .7d0
       ,mnmfor = 25.d0

Regards,

Howard