[daip] UVFIX and MERLIN

[Eric Greisen]

Eduardo Ros writes:
 > Dear Sirs,
 > 
 > We were trying to work with MERLIN data using AIPS, and we have
 > found a problem.  We have to change the coordinates of a MERLIN
 > archive from B1950.0 to J2000.0.  The task appropriate to do
 > that (we thought) is UVFIX.  The problem is that we do not know
 > how UVFIX handles with the UV-data.
 > 

       I do not know what the roblem with UVFIX might be.  It probably
relates to antenna coordinate systems.  If necessary, you can point me
at the Merlin data and I will examine the matter.

    But, to make a data set go from 1950 -> 2000 do the following:

1. Get the Merlin data and type the verb ALTSW

2. KEYWORD = 'CROTAn' where n is the Declination axis in the header
   GETHEAD

2. Task 'UVSRT'; ROTATE = -KEYVAL(1); go

This should make a uv data set with slightly rotated u's and v's and a
header with 0 in the rotation parameter.

Then you may use DBCON.

 > We have applied UVFIX either with parameters UVFIXPRM=0,0, or 1,0, 
 > or even 2,0.  In none of those cases the UV-coverage (say, the 
 > positions of the UV-points corresponding to the different scans) 
 > from the UVFIX output is the same than from the input file.  They
 > are completely different (from each other and from the original
 > one).
 > 
 > A look to the AN-table shows that the antenna positions have not
 > been changed by UVFIX.  Then the change has been done directly in 
 > the (u,v)-data.

      UVFIX is supposed to recompute the U and V based on the time and
antenna positions.  It does not change the AN file or times.


Eric Greisen