[daip] phase rotation

[Eric Greisen]

Andy Biggs writes:
 > Hi there. I'm trying to use UVFIX to move the phase centre of a dataset 
 > and it seems to be working fine as far as I can tell. As usual though, I'm 
 > confused by the UT1-UTC and the IAT-UTC parameters - do these need to be 
 > set? I'm asking because presumably what I'm doing with UVFIX is what IMAGR 
 > does when you use the DO3DIMAG option, but I don't suppose IMAGR knows 
 > what UT1-UTC, etc are. My question directly relates to how I do the shift 
 > when I have a dataset consisting of multiple epochs DBCON'd together. They 
 > may have very different values of UT1-UTC and IAT-UTC.
 > 

You can put the correct values in the headers of each of the AN tables
and then UVFIX will use them rather than the single values in the
input adverbs.  The phase shift is done with the current u,v,w and the
time parameters are used to compute a new u,v,w from the AN tables.
IMAGR recomputes u,v,w with a simple rotation matrix.  Its assumption
is that u,v,w are correct while UVFIX (sort of) assumes the opposite.
This is why there is a recommendation in UVFIX to first run it with no
position shift and then re-run it with the shift.

Eric Greisen