[daip] fitld fix ?

[Lorant Sjouwerman]

Eric, 

This is about the VLA phased array problem.

A short description:

The old way of getting VLA Tsys calibration (and gain curve) is to read in
an auxillary file with ANTAB. ANTAB recognizes whether the VLA is single
dish or phased array from the word "SYS/SRC" in the TSYS card in that file
(ANTAB.FOR line 1147-1149) and sets the TANT values (columns 8 & 10) of the
TY table to -1.0 (line 1170). APCAL then does the right thing, that is for
the phased VLA recognize the TSYS values in the TY table (columns 7 & 9)  
being negative values as being Tsys/Tant (APCAL.FOR line 783) and convert
them to real TSYS values (line 791) for calibration.

Since a year or two we provide the VLA Tsys values (and gain curve) to the
correlator and thus need not use ANTAB to read in an auxillary file from
some web site - the Tsys values are already provided on the distribution
tape and are filled in in the TY table TSYS columns (7 & 9) just as for the
VLBA antennas (and GBT, Eb, Ar). All antennas seem to have the TANT columns
(8 & 10) filled with INDE. For a single dish VLA antenna this is okay as
they are real Tsys values as for the VLBA. For the phased array these
values are TSYS/TANT, but APCAL doesn't recognize these as such because the
TY-table TANT columns (8 & 10) are not specified as -1 (for the VLA rows),
which means that the VLA amplitude calibration with APCAL is bogus.

The VLBA correlator doesn't know about single dish or phased array - it 
just attaches the Tsys ot Tsys/Tant vales it is given in a table.

I think it should be relatively easy (showing my ignorance) to make a fix
for this in FITLD. FITLD would recognize station 'Y' with a corresponding
entry in the provided gain curve (GC) table. FITLD can recognize phased
array by determining if the gain curve is unity (or larger than 0.999), and
if so, only then it would fill VLA TY-columns 8 and 10 with -1 (or .LT.0.0)

I could make a simple PHVLAFIX procedure that uses TABED, which I like to
avoid having users run, or in the worst case we should go back to the user
unfriendly old way (using ANTAB, not have the Tsys in the distribution).


Regards,

Loránt Sjouwerman - Scientific Services - lsjouwerman at nrao.edu
--------------------------------------------------------------
c/o NRAO Array Operations Center       Phone:  +1-505-835-7332
    P.O. Box 0 (1003 Lopezville Rd)    Switch: +1-505-835-7000
    Socorro NM 87801                   Fax:    +1-505-835-7027