general comments:

Antenna 1 is not right - IF1 both polarizations and IF2 R had
reasonable solutions but the gains were a factor 10 high.  IF2 L is
garbage with many failed solutions.

Comments now on phases:

1. Two real jumps associated with scans occurred - both accompanied by
   very large gains and significant changes in Psum and Pdif

    Ant 6  IF 1  04:35:25.5 -> 26.5 (int value) -> 27.5 (new value)
                 04:26:25.5 -> 26.5 (return to old values)
                 note - the end of scan is 25.0

    ant 27 IF 1 same pattern
                 04:14:25.5 -> 26.5 -> 27.5 (new value)
                 04:15:25.5 -> 26.5 (back to ~old value) 

These clearly show that the times recorded with the data are wrong by
one second (or maybe 1.5 or 2) in the last > 1 year of the archive -
unless the scan-related thing is not tied to scans.

2. Three smaller jumps also were found of a smaller and lasting
variety (no return to old values except by smooth atmospheric things)
These were in all IFs and polarizations

        Ant 20  04:39:26.5 -> 27.5  (close to start or start??)
                04:45:26.5 -> 27.5
        Ant 15  05:00:42.5 -> 43.5  (middle of scan)

---------------------------------------------
 
Amplitudes are another question:

There are pervasive scan-based jumps and the SY table does nothing to
correct them when TYAPL is applied.  I attach a PostScript file.
Again the times suggest that we are making a 1 second error in
recorded time.

Eric