[daip] Re: WSRT Multiple IF data in AIPS....

[Eric Greisen]

Nissim Kanekar writes:

 > I've been trying to use AIPS to analyse WSRT data and have run into a 
 > problem when the data contains multiple IFs. The IF selection doesn't seem 
 > to work on many of the tasks (e.g. FITLD, SPLAT, UVCOP, etc). FITLD simply 
 > passes all the IFs in the data through, independent of IF selection, while 
 > SPLAT and UVCOP lose the source names in the output file  if one chooses 
 > IFs other than the 1st. Similarly, BPASS aborts if only one of the IFs is 
 > chosen. And SPLIT produces ExaJy numbers when calibration is applied 
 > to the multi-channel data. On the other hand, AVSPC seemed to work fine. 
 > None of the above problems arise when analysing single IF WSRT data; I 
 > was hence unsure as to whether this is a problem with AIPS or with the 
 > WSRT multi-IF FITS format; could you let me know what you think ? I'm 
 > running the 31DEC02 version (with all patches applied) on Redhat Linux 
 > 7.2, with the GNU 2.95.3 compilers.
 > 

I am unfamiliar with the WSRT data in AIPS.  FITLD applies IF
selection only to VLBA Correlator data in the IDI format.  Plain FITS
files have no data selection done in loading.  However, all the others
should work.

I will need a lot more info.  Probably a good start will be the
IMHEADER of the data set you have just loaded.  A PRTAB of the FQ
and SU tables would also be good.

I am unsure about BPASS's ability to do only one IF - that may be a
more general bug.  Most VLA line data have only 1 IF and users mostly
want all IFs to be calibrated...

The SPLAT, UVCOP, etc suggest that we have a variety of possible
errors in the table formats or something...

Eric Greisen